%cec2014.awk Revision: 1.00 $ 05 Jun 2014 %Carlos A. Coello Coello Wed, Jun 4, 2014 % Special Session: MoE1-1 Computational Intelligence and Games @InProceedings{Lee:2014:CEC, title = {Learning a {Super Mario} Controller from Examples of Human Play}, author = {Geoffrey Lee and Min Luo and Fabio Zambetta and Xiaodong Li}, pages = {1--8}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games, Adaptive dynamic programming and reinforcement learning, Games}, abstract = { Imitating human-like behaviour in action games is a challenging but intriguing task in Artificial Intelligence research, with various strategies being employed to solve the human-like imitation problem. In this research we consider learning human-like behaviour via Markov decision processes without being explicitly given a reward function, and learning to perform the task by observing expert's demonstration. Individual players often have characteristic styles when playing the game, and this method attempts to find the behaviours which make them unique. During play sessions of Super Mario we calculate player's behaviour policies and reward functions by applying inverse reinforcement learning to the player's actions in game. We conduct an online questionnaire which displays two video clips, where one is played by a human expert and the other is played by the designed controller based on the player's policy. We demonstrate that by using apprenticeship learning via Inverse Reinforcement Learning, we are able to get an optimal policy which yields performance close to that of an human expert playing the game, at least under specific conditions. }} @InProceedings{Nguyen:2014:CEC, title = {Integrating Fuzzy Integral and Heuristic Search for Unit Micromanagement in {RTS} Games}, author = {Tung Nguyen and Kien Nguyen and Ruck Thawonmas}, pages = {9--12}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games}, abstract = { Real-time strategy (RTS) is a sub-genre of strategy video game which typically involves resource gathering, base building, strategy planning, and combat scenarios. With complicated gameplay, vast state and action spaces, RTS games have been proved to be an excellent platform for artificial intelligence research. One of the most challenging problems posed by RTS games is the detailed control of units in combat, i.e., unit micromanagement. In this paper, we present a method of integrating fuzzy integral and fast heuristic search for improving the quality of unit micromanagement in the popular RTS game StarCraft. Experiments are reported at the end of this paper, showing promising results and the potential of the proposed method in this domain. }} @InProceedings{Ashlock:2014:CEC, title = {{*Tego} - A Framework for Adversarial Planning}, author = {Daniel Ashlock and Philip Hingston}, pages = {13--20}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games}, abstract = { This study establishes a framework called *-Tego for a situation in which two agents are each given a set of players for a competitive game. Each agent places their players in an order. Players on each side at the same position in the order play one another, with the agent's score being the sum of their player's scores. The planning agents are permitted to simultaneous reorder their players in each of several stages. The reordering is termed competitive replanning. The resulting framework is scalable by changing the number of players and the complexity of the replanning process. The framework is demonstrated using iterated prisoner's dilemma on a set of twenty players. The system is first tested with one agent unable to change the order of its players, yielding an optimisation problem. The system is then tested in a competitive co-evolution of planning agents. The optimisation form of the system makes globally sensible assignments of players. The co-evolutionary version concentrates on matching particular high-payoff pairs of players with the agents repeatedly reversing one another's assignments, with the majority of players with smaller payoffs at risk are largely ignored. }} @InProceedings{Gaudesi:2014:CEC, title = {{TURAN}: Evolving Non-Deterministic Players for the Iterated Prisoner's Dilemma}, author = {Marco Gaudesi and Elio Piccolo and Giovanni Squillero and Alberto Tonda}, pages = {21--27}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games, Games, Evolutionary games and multi-agent systems}, abstract = { The iterated prisoner's dilemma is a widely known model in game theory, fundamental to many theories of co-operation and trust among self-interested beings. There are many works in literature about developing efficient strategies for this problem, both inside and outside the machine learning community. This paper shift the focus from finding a "good strategy" in absolute terms, to dynamically adapting and optimising the strategy against the current opponent. Turan evolves competitive non-deterministic models of the current opponent, and exploit them to predict its moves and maximise the payoff as the game develops. Experimental results show that the proposed approach is able to obtain good performances against different kind of opponent, whether their strategies can or cannot be implemented as finite state machines. }} @InProceedings{Buck:2014:CEC, title = {Evolving a Fuzzy Goal-Driven Strategy for the Game of {Geister}}, author = {Andrew Buck and Tanvi Banerjee and James Keller}, pages = {28--35}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games, Evolved neuro-fuzzy systems, Games}, abstract = { This paper presents an approach to designing a strategy for the game of Geister using the three main research areas of computational intelligence. We use a goal-based fuzzy inference system to evaluate the utility of possible actions and a neural network to estimate unobservable features (the true natures of the opponent ghosts). Finally, we develop a coevolutionary algorithm to learn the parameters of the strategy. The resulting autonomous gameplay agent was entered in a global competition sponsored by the IEEE Computational Intelligence Society and finished second among eight participating teams. }} @InProceedings{Handa:2014:CEC, title = {Deep Boltzmann Machine for Evolutionary Agents of {Mario AI}}, author = {Hisashi Handa}, pages = {36--41}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games, Games}, abstract = { Deep Learning has attracted much attention recently since it can extract features taking account into the high-order knowledge. In this paper, we examine the Deep Boltzmann Machines for scene information of the Mario AI Championship. That is, the proposed method is composed of two parts: the DBM and a recurrent neural network. The DBM extracts features behind perceptual scene information, and it learns off-line. On the other hand, the recurrent neural network uses features to decide actions of the Mario AI agents, and it learns on-line by using Particle Swarm Optimisation. Experimental results show the effectiveness of the proposed method. }} % Special Session: MoE1-2 Memetic Computing @InProceedings{Rahman:2014:CEC, title = {A Memetic Algorithm for Solving Permutation Flow Shop Problems with Known and Unknown Machine Breakdowns}, author = {Humyun Fuad Rahman and Ruhul Sarker and Daryl Essam and Guijuan Chang}, pages = {42--49}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing}, abstract = { The Permutation Flow Shop Scheduling Problem (PFSP) is considered to be one of the complex combinatorial optimisation problems. For PFSPs, the schedule is produced under ideal conditions that usually ignore any type of process interruption. In practice, the production process is interrupted due to many different reasons, such as machine unavailability and breakdowns. In this paper, we propose a Genetic Algorithm (GA) based approach to deal with process interruptions at different points in time in Permutation Shop Floor scenarios. We have considered two types of process interruption events. The first one is predictive, where the interruption information is known well in advance, and the second one is reactive, where the interruption information is not known until the breakdown occurs. An extensive set of experiments has been carried out, which demonstrate the usefulness of the proposed approach. }} @InProceedings{Ma:2014:CEC, title = {Remote Sensing Imagery Clustering Using an Adaptive Bi-Objective Memetic Method}, author = {Ailong Ma and Yanfei Zhong and Liangpei Zhang}, pages = {50--57}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing}, abstract = { Due to the intrinsic complexity of the remote sensing image and the lack of the prior knowledge, clustering for remote sensing image has always been one of the most challenging works in remote sensing image processing. The proposed algorithm constructs a bi-objective memetic-based framework, exploiting the feature space more efficiently. In the framework, two objective functions, Jm and XB, are used as the objective functions for bi-objective optimisation. Furthermore, an adaptive local search method which can dynamically adjust its parameter value according to the selection probability has been developed and incorporated into the proposed algorithm. In order to speed the convergence and obtain more non-dominated solutions in the Pareto front, a new strategy is newly devised in the local search process, which considers more solutions as the candidate for the next generation. To evaluate the proposed algorithm, some experiments on two multi-spectral images are conducted. The results show that the proposed algorithm can achieve better performance, compared with related methods. }} @InProceedings{Ma:2014:CECa, title = {A Memetic Algorithm Based on Immune Multi-Objective Optimization for Flexible Job-Shop Scheduling Problems}, author = {Jingjing Ma and Yu Lei and Zhao Wang and Licheng Jiao}, pages = {58--65}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing}, abstract = { The flexible job-shop scheduling problem (FJSP) is an extension of the classical job scheduling which is concerned with the determination of a sequence of jobs, consisting of many operations, on different machines, satisfying parallel goals. This paper addresses the FJSP with two objectives: Minimise makespan, Minimise total operation cost. We introduce a memetic algorithm based on the Nondominated Neighbour Immune Algorithm (NNIA), to tackle this problem. The proposed algorithm adds, to NNIA, local search procedures including a rational combination of undirected simulated annealing (UDSA) operator, directed cost simulated annealing (DCSA) operator and directed makespan simulated annealing (DMSA) operator. We have validated its efficiency by evaluating the algorithm on multiple instances of the FJSPs. Experimental results show that the proposed algorithm is an efficient and effective algorithm for the FJSPs, and the combination of UDSA operator, DCSA operator and DMSA operator with NNIA is rational. }} @InProceedings{Ma:2014:CECb, title = {A Memetic Algorithm for Solving Flexible Job-Shop Scheduling Problems}, author = {Wenping Ma and Yi Zuo and Jiulin Zeng and Shuang Liang and Licheng Jiao}, pages = {66--73}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing}, abstract = { The flexible Job-shop Scheduling Problem (FJSP) is an extension of the classical job-shop scheduling problem (JSP). In this paper, a memetic algorithm (MA) for the FJSP is presented. This MA is a hybrid genetic algorithm which explores the search space and two efficient local searchers to exploit information in the search region. An extensive computational study on 49 benchmark problems shows that the algorithm is effective and robust, with respect to other well-known effective algorithms. }} @InProceedings{Wei:2014:CEC, title = {Hybridizing the Dynamic Mutation Approach with Local Searches to Overcome Local Optima}, author = {Kuai Wei and Michael J. Dinneen}, pages = {74--81}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing, Evolutionary computation theory}, abstract = { A Memetic Algorithm is an Evolutionary Algorithm augmented with local searches. The dynamic mutation approach has been studied extensively in experiments of Memetic Algorithms, but only a few studies in theory. We previously defined a metric BLOCKONES to estimate the difficulty of escaping from a local optima. An algorithm's ability of escaping from a local optima that has a large BLOCKONES is very important because it dominates the time complexity of finding a global optimal solution. In this paper, we will use the same metric and show the benefits of hybridising the dynamic mutation approach with one of two local searches, best-improvement and first-improvement. In short, this hybridisation greatly enhances the algorithm's ability to escape from any local optima. }} @InProceedings{Liu:2014:CEC, title = {Memetic Algorithm with Adaptive Local Search Depth for Large Scale Global Optimization}, author = {Can Liu and Bin Li}, pages = {82--88}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic Computing}, abstract = { Memetic algorithms (MAs) have been recognised as an effective algorithm framework for solving optimisation problems. However, the exiting work mainly focused on the improvement for search operators. Local Search Depth (LSD) is a crucial parameter in MAs, which controls the computing resources assigned for local search. In this paper, an Adaptive Local Search Depth (ALSD) strategy is proposed to arrange the computing resources for local search according to its performance dynamically. A Memetic Algorithm with ALSD (MA-ALSD) is presented, its performance and the effectiveness of ALSD are testified via experiments on the LSGO test suite issued in CEC'2012. }} % Special Session: MoE1-3 Evolutionary Computer Vision @InProceedings{Albukhanajer:2014:CEC, title = {Neural Network Ensembles for Image Identification Using {Pareto}-Optimal Features}, author = {Wissam A. Albukhanajer and Yaochu Jin and Johann A. Briffa}, pages = {89--96}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computer Vision , Real-world applications, Multi-objective evolutionary algorithms}, abstract = { In this paper, an ensemble classifier is constructed for invariant image identification, where the inputs to the ensemble members are a set of Pareto-optimal image features extracted by an evolutionary multi-objective Trace transform algorithm. The Pareto-optimal feature set, called Triple features, gains various degrees of trade-off between sensitivity and invariance. Multilayer perceptron neural networks are adopted as ensemble members due to their simplicity and capability for pattern classification. The diversity of the ensemble is mainly achieved by the Pareto-optimal features extracted by the multi-objective evolutionary Trace transform. Empirical results show that the general performance of proposed ensemble classifiers is more robust to geometric deformations and noise in images compared to single neural network classifiers using one image feature. }} @InProceedings{Valsecchi:2014:CEC, title = {Automatic Evolutionary Medical Image Segmentation Using Deformable Models}, author = {Andrea Valsecchi and Pablo Mesejo and Linda Marrakchi-Kacem and Stefano Cagnoni and Sergio Damas}, pages = {97--104}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computer Vision }, abstract = { This paper describes a hybrid level set approach to medical image segmentation. The method combines region and edge based information with the prior shape knowledge introduced using deformable registration. A parameter tuning mechanism, based on Genetic Algorithms, provides the ability to automatically adapt the level set to different segmentation tasks. Provided with a set of examples, the GA learns the correct weights for each image feature used in the segmentation. The algorithm has been tested over four different medical datasets across three image modalities. Our approach has shown significantly more accurate results in comparison with six state-of-the-art segmentation methods. The contribution of both the image registration and the parameter learning steps to the overall performance of the method has also been analysed. }} @InProceedings{Schaefer:2014:CEC, title = {Cost-Sensitive Texture Classification}, author = {Gerald Schaefer and Bartosz Krawczyk and Niraj Doshi and Tomoharu Nakashima}, pages = {105--108}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computer Vision }, abstract = { Texture recognition plays an important role in many computer vision tasks including segmentation, scene understanding and interpretation, medical imaging and object recognition. In some situations, the correct identification of particular textures is more important compared to others, for example recognition of enemy uniforms for automatic defence systems, or isolation of textures related to tumours in medical images. Such cost-sensitive texture classification is the focus of this paper, which we address by reformulating the classification problem as a cost minimisation problem. We do this by constructing a cost-sensitive classifier ensemble that is tuned using a genetic algorithm. Based on experimental results obtained on several Outex datasets with cost definitions, we show our approach to work well in comparison with canonical classification methods and the ensemble approach to lead to better performance compared to single predictors. }} @InProceedings{Naqvi:2014:CEC, title = {Genetic Algorithms Based Feature Combination for Salient Object Detection, for Autonomously Identified Image Domain Types}, author = {Syed Saud Naqvi and Will N. Browne and Christopher Hollitt}, pages = {109--116}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computer Vision }, abstract = { Combining features from different modalities and domains has been demonstrated to enhance the performance of saliency prediction algorithms. Different feature combinations are often suited to different types of images, but existing techniques attempt to apply a single feature combination across all image types. Furthermore, existing normalisation and integration schemes are not used in salient object detection as the combination of potential solutions is intractable to test. The aim of this work is to autonomously learn feature combinations for autonomously identified image types. To this end, we learn optimal normalisation and integration schemes along with feature weightings using a novel Genetic Algorithm (GA) method. Moreover, we learn multiple image dependent parameters using our novel image-based GA (IGA) approach, to increase the generalisation of the system on unseen test images. We present a thorough quantitative and qualitative comparison of our proposed methods with the state-of-the-art benchmark and deterministic methods on two difficult datasets (SED1 and SED2) from the segmentation evaluation database. IGA shows superior performance through learning optimal parameters depending upon the composition of images and using feature combinations appropriately enhances test performance and generalisation of the system. }} @InProceedings{Fu:2014:CEC, title = {Unsupervised Learning for Edge Detection Using Genetic Programming}, author = {Wenlong Fu and Mark Johnston and Mengjie Zhang}, pages = {117--124}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Evolutionary Computer Vision }, abstract = { In edge detection, a machine learning algorithm generally requires training images with their ground truth or designed outputs to train an edge detector. Meanwhile the computational cost is heavy for most supervised learning algorithms in the training stage when a large set of training images is used. To learn edge detectors without ground truth and reduce the computational cost, an unsupervised Genetic Programming (GP) system is proposed for low-level edge detection. A new fitness function is developed from the energy functions in active contours. The proposed GP system utilises single images to evolve GP edge detectors, and these evolved edge detectors are used to detect edges on a large set of test images. The results of the experiments show that the proposed unsupervised learning GP system can effectively evolve good edge detectors to quickly detect edges on different natural images. }} % Special Session: MoE1-4 Theoretical Foundations of Bio-inspired Computation @InProceedings{Wagner:2014:CEC, title = {Single- and Multi-Objective Genetic Programming: New Runtime Results for {SORTING}}, author = {Markus Wagner and Frank Neumann}, pages = {125--132}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Theoretical Foundations of Bio-inspired Computation}, abstract = { In genetic programming, the size of a solution is typically not specified in advance and solutions of larger size may have a larger benefit. The flexibility often comes at the cost of the so-called bloat problem: individuals grow without providing additional benefit to the quality of solutions, and the additional elements can block the optimisation process. Consequently, problems that are relatively easy to optimise cannot be handled by variable-length evolutionary algorithms. In this article, we present several new bounds for different single and multi-objective algorithms on the sorting problem, a problem that typically lacks independent and additive fitness structures. }} @InProceedings{Wei:2014:CECa, title = {Runtime Comparison of Two Fitness Functions on a Memetic Algorithm for the Clique Problem}, author = {Kuai Wei and Michael J. Dinneen}, pages = {133--140}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Theoretical Foundations of Bio-inspired Computation, Memetic, multi-meme and hybrid algorithms}, abstract = { It is commonly accepted that a proper fitness function can guide the algorithm to find a global optimum solution faster. This paper will use the runtime analysis to provide the theoretical evidence that a small change of the fitness function can result in a huge performance gap in terms of finding a global optimum solution. It also shows that the fitness function that gives the best results in an Memetic Algorithm on the Clique Problem is entirely instance specific. In detail, we will formalise a (1+1) Restart Memetic Algorithm with a Random Complete Local Search, and run them on two different fitness functions, f\_OL and f\_OPL, to solve the Clique Problem respectively. We then construct two families of graphs, G\_1 and G\_2, and show that, for the first family of graphs G\_1, the (1+1) RMA on the fitness function f\_OPL drastically outperforms the (1+1) RMA on the fitness function f\_OL, and vice versa for the second family of graphs G\_2. }} @InProceedings{He:2014:CEC, title = {A Theoretical Assessment of Solution Quality in Evolutionary Algorithms for the Knapsack Problem}, author = {Jun He and Mitavskiy Boris and Yuren Zhou}, pages = {141--148}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Theoretical Foundations of Bio-inspired Computation, Discrete and combinatorial optimisation, Multi-objective evolutionary algorithms}, abstract = { Evolutionary algorithms are well suited for solving the knapsack problem. Some empirical studies claim that evolutionary algorithms can produce good solutions to the 0-1 knapsack problem. Nonetheless, few rigorous investigations address the quality of solutions that evolutionary algorithms may produce for the knapsack problem. This paper focuses on a theoretical investigation of three types of (N+1) evolutionary algorithms that exploit bitwise mutation, truncation selection, plus different repair methods for the 0-1 knapsack problem. It assesses the solution quality in terms of the approximation ratio. Our work indicates that the solution produced by both pure strategy and mixed strategy evolutionary algorithms is arbitrarily bad. Nevertheless, an evolutionary algorithm using helper objectives may produce 1/2-approximation solutions to the 0-1 knapsack problem. }} @InProceedings{Yu:2014:CEC, title = {The Sampling-and-Learning Framework: A Statistical View of Evolutionary Algorithms}, author = {Yang Yu and Hong Qian}, pages = {149--158}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Theoretical Foundations of Bio-inspired Computation, Evolutionary computation theory}, abstract = { Evolutionary algorithms (EAs), a large class of general purpose optimisation algorithms inspired from the natural phenomena, are widely used in various industrial optimisations and often show excellent performance. This paper presents an attempt towards revealing their general power from a statistical view of EAs. By summarising a large range of EAs into the sampling-and-learning framework, we show that the framework directly admits a general analysis on the probable-absolute-approximate (PAA) query complexity. We particularly focus on the framework with the learning subroutine being restricted as a binary classification, which results in the sampling-and-classification (SAC) algorithms. With the help of the learning theory, we obtain a general upper bound on the PAA query complexity of SAC algorithms. We further compare SAC algorithms with the uniform search in different situations. Under the error-target independence condition, we show that SAC algorithms can achieve polynomial speedup to the uniform search, but not super-polynomial speedup. Under the one-side-error condition, we show that super-polynomial speedup can be achieved. This work only touches the surface of the framework. Its power under other conditions is still open. }} @InProceedings{Chotard:2014:CEC, title = {Markov Chain Analysis of Evolution Strategies on a Linear Constraint Optimization Problem}, author = {Alexandre Chotard and Anne Auger and Nikolaus Hansen}, pages = {159--166}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Theoretical Foundations of Bio-inspired Computation, Convergence, scalability and complexity analysis, Evolution strategies}, abstract = { This paper analyses a \$(1,{$\backslash$}lambda)\$-Evolution Strategy, a randomised comparison-based adaptive search algorithm, on a simple constraint optimisation problem. The algorithm uses re-sampling to handle the constraint and optimises a linear function with a linear constraint. Two cases are investigated: first the case where the step-size is constant, and second the case where the step-size is adapted using path length control. We exhibit for each case a Markov chain whose stability analysis would allow us to deduce the divergence of the algorithm depending on its internal parameters. We show divergence at a constant rate when the step-size is constant. We sketch that with step-size adaptation geometric divergence takes place. Our results complement previous studies where stability was assumed. }} @InProceedings{Everitt:2014:CEC, title = {Free Lunch for Optimisation under the Universal Distribution}, author = {Tom Everitt and Tor Lattimore and Marcus Hutter}, pages = {167--174}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial optimisation, Evolutionary computation theory, Convergence, scalability and complexity analysis}, abstract = { Function optimisation is a major challenge in computer science. The No Free Lunch theorems state that if all functions with the same histogram are assumed to be equally probable then no algorithm outperforms any other in expectation. We argue against the uniform assumption and suggest a universal prior exists for which there is a free lunch, but where no particular class of functions is favoured over another. We also prove upper and lower bounds on the size of the free lunch. }} % Plenary Poster Session: PE1 Poster Session I @InProceedings{Arana-Daniel:2014:CEC, title = {Smooth Global and Local Path Planning for Mobile Robot Using Particle Swarm Optimization, Radial Basis Functions, Splines and {Bezier} Curves}, author = {Nancy Arana-Daniel and Alberto A. Gallegos and Carlos Lopez-Franco and Alma Y. Alanis}, pages = {175--182}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Robotics}, abstract = { An approach to plan smooth paths for mobile robots using a Radial Basis Function (RBF) neural network trained with Particle Swarm Optimisation (PSO) was presented before by the authors in [1]. Taking the previous approach as an starting point, in this paper it is shown that it is possible to construct a smooth simple global path and then modify this path locally using PSO-RBF, Ferguson splines or Bezier curves trained with PSO, in order to describe more complex paths or to deal with dynamic changes in the environment. Experimental results show that our approach is fast and effective to deal with complex environments. }} @InProceedings{Wang:2014:CEC, title = {A Novel Improvement of Particle Swarm Optimization Using Dual Factors Strategy}, author = {Lin Wang and Bo Yang and Yi Li and Na Zhang}, pages = {183--189}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Genetic algorithms}, abstract = { The particle swarm optimisation, inspired by nature, is widely used for optimising complex problems and achieves many good stories in practical applications. However, the traditional PSO only focuses on the function value during evolutionary process. It ignores the information of distance between particles and potential regions. A Dual Factors Particle Swarm Optimisation (DFPSO) incorporating both of distance and function information is proposed in this paper to help PSO in finding potential global optimal regions. The strategy of the DFPSO increases the diversity of population to yield improved results. The experimental results manifest that the performance, including accuracy and speed, are improved. }} @InProceedings{Xiang:2014:CEC, title = {A Verifiable {PSO} Algorithm in Cloud Computing}, author = {Tao Xiang and Weimin Zhang and Fei Chen}, pages = {190--193}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Defence and cyber security}, abstract = { In this paper, we study the verification problem of particle swarm optimisation (PSO) when it is outsourced to the cloud, i.e. making sure that the cloud executes PSO algorithm as requested. A verifiable PSO algorithm and its verification algorithm are proposed. The proposed scheme does not involve expensive cryptography, and it is efficient and effective to verify the honesty of the cloud. }} @InProceedings{Zong:2014:CEC, title = {Space-Time Simulation Model Based on Particle Swarm Optimization Algorithm for Stadium Evacuation}, author = {Xinlu Zong and Shengwu Xiong and Hui Xu and Pengfei Duan}, pages = {194--201}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Intelligent systems applications, Evolutionary Computing with Deterministic Chaos}, abstract = { In this paper, a space-time simulation model based on particle swarm optimisation algorithm for stadium evacuation is presented. In this new model, the fast evacuation, going with the crowd and the panic behaviours are considered and the corresponding moving rules are defined. The model is applied to a stadium and simulations are carried out to analyse the space-time evacuation efficiency by different behaviours. The simulation results show that the behaviours of going with the crowd and panic will slow down the evacuation process while quickest evacuation psychology can accelerate the process, and panic is helpful to some extent. The setting of parameters is discussed to obtain best performance. The simulation results can offer effective suggestions for evacuees under emergency situation. }} @InProceedings{Campos:2014:CEC, title = {Bare Bones Particle Swarm with Scale Mixtures of {Gaussians} for Dynamic Constrained Optimization}, author = {Mauro Campos and Renato Krohling}, pages = {202--209}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { Bare bones particle swarm optimisation (BBPSO) is a well-known swarm algorithm which has shown potential for solving single-objective constrained optimisation problems in static environments. In this paper, a generalised BBPSO for dynamic single-objective constrained optimisation problems is proposed. An empirical study was carried out to evaluate the performance of the proposed approach. Experimental results show the suitability of the proposed algorithm in terms of effectiveness to find good solutions for all benchmark problems investigated. For comparison purposes, experimental results found by other algorithms are also presented. }} @InProceedings{Zhang:2014:CEC, title = {Cooperative Particle Swarm Optimizer with Elimination Mechanism for Global Optimization of Multimodal Problems}, author = {Geng Zhang and Yangmin Li}, pages = {210--217}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { This paper presents a new particle swarm optimiser (PSO) that called the cooperative particle swarm optimiser with elimination mechanism (CPSO-EM) in an attempt to address the issue of getting trapped into local optimum when solving nonseparable multimodal problems using PSO algorithm. The proposed CPSO-EM builds on the basis of an early cooperative PSO (CPSO-H) that employs cooperative behaviour. The CPSOH and elimination mechanism (EM) memory are incorporated together to obtain CPSO-EM. Experimental studies on a set of test functions show that CPSO-EM exhibits better performance in solving nonseparable multimodal problems than several other peer algorithms. }} @InProceedings{Yan:2014:CEC, title = {A Chaotic Particle Swarm Optimization Algorithm for the Jobshop Scheduling Problem}, author = {Ping Yan and Minghai Jiao}, pages = {218--222}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO)}, abstract = { An improved Chaotic Particle Swarm Optimisation (CPSO) algorithm for a jobshop scheduling problem, with minimisation of makespan as the criterion, is proposed in this research. A real-valued encoding scheme based on a matrix representation is developed, which converts the continuous position value of particles in PSO to the processing order of job operation. A compound chaotic search strategy that integrates both Tent and Logistic chaotic search process is employed to the global best particle to enhance the local searching ability of PSO. In addition, a Gaussian disturbance technology is embedded in the CPSO algorithm to improve the diversity of the particles in the swarm. The performance of CPSO is compared with the standard PSO algorithm on a benchmark instance of jobshop scheduling problems. The results show that the proposed CPSO algorithm has a superior performance to the PSO algorithm. }} @InProceedings{Dong:2014:CEC, title = {Autonomous Learning Adaptation for Particle Swarm Optimization}, author = {Wenyong Dong and Jiangshen Tian and Xu Tang and Kang Sheng and Jin Liu}, pages = {223--228}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Self-adaptation in evolutionary computation, Particle swarm optimisation (PSO)}, abstract = { In order to improve the performance of PSO, this paper presents an Autonomous Learning Adaptation method for Particle Swarm Optimisation (ALA-PSO) to automatically tune the control parameters of each particle. Although PSO is an ideal optimiser, one of its drawbacks focuses on its performance dependency on its parameters, which differ from one problem to another. In ALA-PSO, each particle is viewed as an intelligent agent and aim at improve itself performance, and can autonomously learn how to tune its parameters from its own experiment of successes and failures. For each particle, it means success movement if the value of objective function in current position is improved than previous position, otherwise means failure. In case of successful movement, the parameters that are positive correlation with the direction of forward movement should be increased otherwise should be decreased. Meanwhile, in case of unsuccessful movement, inverse operation should be performed. The proposed parameter adaptive method is compared with several existing adaptive strategies, and the results show that ALA-PSO is not only effective, but also robust in different categories benchmarks. }} @InProceedings{Wu:2014:CEC, title = {A Growing Partitional Clustering Based on Particle Swarm Optimization}, author = {Nuosi Wu and Zexuan Zhu and Zhen Ji}, pages = {229--234}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Data Mining and Machine Learning Meet Evolutionary Computation, Classification, clustering and data analysis, Particle swarm optimisation (PSO)}, abstract = { This paper proposes a growing partitional clustering method based on particle swarm optimisation (PSO) namely PSOGC for handling data with non-spherical or non-linearly separable distribution. Particularly, PSOGC uses PSO to optimise the cluster centres. In each iteration of PSO, the particles encoding candidate cluster centres are evolved according to their social and personal knowledge. Given the candidate cluster centres, a growing strategy increasingly absorbs nearby data samples into the corresponding cluster based on k-nearest neighbour graph. The fitness of each particle is evaluated in terms of intra-cluster connectivity and inter-cluster disconnectivity of the resultant clustering. The combination of PSO and growing strategy ensures the stability of global search and the robustness of partition on data of different non-spherical shapes. Experimental results on six synthetic and three UCI real-world data sets demonstrate the efficiency of PSOGC. }} @InProceedings{Kuang:2014:CEC, title = {A Novel Chaotic Artificial Bee Colony Algorithm Based on Tent Map}, author = {Fangjun Kuang and Zhong Jin and Weihong Xu and Siyang Zhang}, pages = {235--241}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Heuristics, metaheuristics and hyper-heuristics, Evolutionary computation theory}, abstract = { A novel self-adaptive chaotic artificial bee colony algorithm based on Tent map (STOC-ABC) is proposed to enhance the global convergence and the population diversity. In the STOC-ABC, Tent chaotic opposition-based learning initialisation method is presented to diversify the initial individuals and obtain good initial solutions. Furthermore, the self-adaptive Tent chaotic searching is implemented at the zones nearby individual optimum solution to help the artificial bee colony (ABC) algorithm to escape from the local optimum effectively. Moreover, the tournament selection strategy in onlooker bee phase is employed to increase the ability of the algorithm and avoid premature convergence. Experiments on six complex benchmark functions with high-dimension, the results further demonstrate that, the STOC-ABC not only accelerates the convergence rate and improves solution precision, but also provides excellent performance in dealing with complex high dimensional functions. }} @InProceedings{Chen:2014:CEC, title = {A Novel Artificial Bee Colony Algorithm with Integration of Extremal Optimization for Numerical Optimization Problems}, author = {Min-Rong Chen and Wei Zeng and Guo-Qiang Zeng and Xia Li and Jian-Ping Luo}, pages = {242--249}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Coevolution and collective behaviour, Heuristics, metaheuristics and hyper-heuristics}, abstract = { Artificial Bee Colony (ABC) algorithm is an optimisation algorithm based on a particular intelligent behaviour of honeybee swarms. The standard artificial bee algorithm is weak at the locally searching capability and precision. Extremal Optimisation (EO) is a general-purpose heuristic method which has strong local search capability and has been successfully applied to a wide variety of hard optimisation problems. In order to strengthen the local-search capability of ABC, this work proposes a novel hybrid algorithm, called ABC-EO algorithm, through introducing EO to ABC. The simulation results show that the performance of the proposed algorithm is superior to those of the state-of-the-art algorithms in complex numerical optimisation problems. }} @InProceedings{Lauri:2014:CEC, title = {Hybrid {ACO/EA} Algorithms Applied to the Multi-Agent Patrolling Problem}, author = {Fabrice Lauri and Abder Koukam}, pages = {250--257}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation, Genetic algorithms, Parallel and distributed algorithms}, abstract = { Patrolling an environment consists in visiting as frequently as possible its most relevant areas in order to supervise, control or protect it. This task is commonly performed by a team of agents that need to coordinate their actions for achieving optimal performance. We address here the problem of multi-agent patrolling in known environments where agents may move at different speeds and visit priorities on some areas may be specified. Two classes of patrolling strategies are studied: the single-cycle strategies and the partition-based strategies. Several single-core and multi-core variants of a template state-of-the-art hybrid algorithm are proposed for generating partition-based strategies. These are experimentally compared with a state-of-the-art heuristic-based algorithm generating single-cycle strategies. Experimental results show that: the heuristic-based algorithm only generates efficient strategies when agents move at the same speeds and no visit priorities have been defined; all single-core variants are equivalent; multi-core hybrid algorithms may improve overall quality or reduce variance of the solutions obtained by single-core algorithms. }} @InProceedings{Zeng:2014:CEC, title = {Comparison of Multiobjective Particle Swarm Optimization and Evolutionary Algorithms for Optimal Reactive Power Dispatch Problem}, author = {Yujiao Zeng and Yanguang Sun}, pages = {258--265}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Multiobjective optimisation, Engineering applications, Multi-objective evolutionary algorithms}, abstract = { The optimal reactive power dispatch (ORPD) problem is formulated as a complex multiobjective optimisation problem, involving nonlinear functions, continuous and discrete variables and various constraints. Recently, multiobjective evolutionary algorithms (MOEAs) and multiobjective particle swarm optimisation (MOPSO) have received a growing interest in solving the multiobjective optimisation problems. In this paper, MOPSO, and two highly competitive algorithms of MOEAs, that is, nondominated sorting genetic algorithm II (NSGA-II) and strength Pareto evolutionary algorithm (SPEA2) are presented for solving the ORPD problem. Moreover, a mixed-variable handling method and an effective constraint handling approach are employed to deal with various types of variables and constraints. The proposed algorithms are evaluated on the standard IEEE 30-bus and 118-bus test systems. In addition, several multiobjective performance metrics are employed to compare these algorithms with respect to convergence, diversity, and computational efficiency. The results show the effectiveness of MOEAs and MOPSO for solving the ORPD problem. Furthermore, the comparison results indicate that MOPSO generally outperforms other algorithms for ORPD and has a great potential in dealing with large-scale optimal power flow problems. }} @InProceedings{Chaman-Garcia:2014:CEC, title = {{MOPSOhv}: A New Hypervolume-Based Multi-Objective Particle Swarm Optimizer}, author = {Ivan Chaman-Garcia and Carlos Coello Coello and Alfredo Arias-Montano}, pages = {266--273}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Multiobjective optimisation, Multi-objective evolutionary algorithms}, abstract = { This paper proposes a new hypervolume-based multi-objective particle swarm optimiser (called MOPSOhv) that uses an external archive to store the global nondominated solutions found during the evolutionary process. The proposed algorithm makes use of the hypervolume contribution of archived solutions for selecting global and personal leaders for each particle in the main swarm, and also as a mechanism for pruning the external archive when it is updated with new nondominated solutions. In order to increase the diversity when particles are updated in their positions, a mutation operator is used. The performance of the proposed algorithm is evaluated adopting standard test problems and indicators reported in the specialised literature, comparing its results with respect to those obtained by state-of-the-art multi-objective evolutionary algorithms. Our preliminary results indicate that our proposal is competitive with respect to state-of-the-art multi-objective evolutionary algorithms, being particularly suitable for solving many-objective optimisation problems (i.e., problems having more than 3 objectives). }} @InProceedings{Peng:2014:CEC, title = {A Population Diversity Maintaining Strategy Based on Dynamic Environment Evolutionary Model for Dynamic Multiobjective Optimization}, author = {Zhou Peng and Jinhua Zheng and Juan Zou}, pages = {274--281}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic Multi-objective Optimisation}, abstract = { Maintaining population diversity is a crucial issue for the performance of dynamic multiobjective optimisation algorithms. However traditional dynamic multiobjective evolutionary algorithms usually imitate the biological evolution of their own, maintain population diversity through different strategies and make the population be able to track the Pareto optimal solution set after the change efficiently. Nevertheless, these algorithms neglect the role of dynamic environment in evolution, lead to the lacking of active and instructional search. In this paper, a population diversity maintaining strategy based on dynamic environment evolutionary model is proposed (DEE-PDMS). This strategy builds a dynamic environment evolutionary model when a change is detected, which makes use of the dynamic environment to record the different knowledge and information generated by population before and after environmental change, and in turn the knowledge and information guide the search in new environment. The model enhances population diversity by guided fashion, makes the simultaneous evolution of the environment and population. A comparison study with other two state-of-the-art strategies on five test problems with linear or nonlinear correlation between design variables has shown the effectiveness of the DEE-PDMS for dealing with dynamic environments. }} @InProceedings{Carvalho:2014:CEC, title = {Multi-Objective Flexible Job-Shop Scheduling Problem with {DIPSO}: More Diversity, Greater Efficiency}, author = {Luiz Carvalho and Marcia Fernandes}, pages = {282--289}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Multi-objective evolutionary algorithms, Genetic algorithms}, abstract = { The Flexible Job Shop Problem is one of the most important NP-hard combinatorial optimisation problems. Evolutionary computation has been widely used in research concerning this problem due to its ability for dealing with large search spaces and the possibility to optimise multiple objectives. Particle Swarm Optimisation has presented good results but the algorithms based on this technique have premature convergence, therefore some proposals introduce genetic operators or other local search methods in order to avoid the local minimum. Therefore, this paper presents a hybrid and multi-objective algorithm, Particle Swarm Optimisation with Diversity (DIPSO), based on Particle Swarm Optimisation along with the genetic operators and Fast Non-dominated Sorting. Thus, to maintain a high degree of diversity in order to guide the search for a better solution while ensuring convergence, a new crossover operator is introduced. The efficiency of this operator is tested in relation to the proposed objectives by using typical examples from the literature. The results are compared to other studies that have had good results by means some Evolutionary Computation technique as for instance MOEAGLS, MOGA, PSO + SA and PSO + TS. }} @InProceedings{Hu:2014:CEC, title = {Calculating the Complete {Pareto} Front for a Special Class of Continuous Multi-Objective Optimization Problems}, author = {Xiao-Bing Hu and Ming Wang and Mark S Leeson}, pages = {290--297}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Hybrid evolutionary Computational Methods for Complex Optimisation Problems}, abstract = { Existing methods for multi-objective optimisation usually provide only an approximation of a Pareto front, and there is little theoretical guarantee of finding the real Pareto front. This paper is concerned with the possibility of fully determining the true Pareto front for those continuous multi-objective optimisation problems for which there are a finite number of local optima in terms of each single objective function and there is an effective method to find all such local optima. To this end, some generalised theoretical conditions are firstly given to guarantee a complete cover of the actual Pareto front for both discrete and continuous problems. Then based on such conditions, an effective search procedure inspired by the rising sea level phenomenon is proposed particularly for continuous problems of the concerned class. Even for general continuous problems to which not all local optima are available, the new method may still work well to approximate the true Pareto front. The good practicability of the proposed method is especially underpinned by multi-optima evolutionary algorithms. The advantages of the proposed method in terms of both solution quality and computational efficiency are illustrated by the simulation results. }} @InProceedings{Lara-Cabrera:2014:CEC, title = {A Self-Adaptive Evolutionary Approach to the Evolution of Aesthetic Maps for a {RTS} Game}, author = {Raul Lara-Cabrera and Carlos Cotta and Antonio J. Fernandez-Leiva}, pages = {298--304}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games}, abstract = { Procedural content generation (PCG) is a research field on the rise, with numerous papers devoted to this topic. This paper presents a PCG method based on a self-adaptive evolution strategy for the automatic generation of maps for the real-time strategy (RTS) game Planet Wars. These maps are generated in order to fulfil the aesthetic preferences of the user, as implied by her assessment of a collection of maps used as training set. A topological approach is used for the characterisation of the maps and their subsequent evaluation: the sphere-of-influence graph (SIG) of each map is built, several graph-theoretic measures are computed on it, and a feature selection method is used to determine adequate subsets of measures to capture the class of the map. A multiobjective evolutionary algorithm is subsequently employed to evolve maps, using these feature sets in order to measure distance to good (aesthetic) and bad (non-aesthetic) maps in the training set. The so-obtained results are visually analysed and compared to the target maps using a Kohonen network. }} @InProceedings{Cai:2014:CEC, title = {Enhanced Differential Evolution with Adaptive Direction Information}, author = {Yiqiao Cai and Jixiang Du}, pages = {305--312}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques}, abstract = { Most recently, a DE framework with neighbourhood and direction information (NDi-DE) was proposed to exploit the information of population and was demonstrated to be effective for most of the DE variants. However, the performance of NDi-DE heavily depends on the selection of direction information. In order to alleviate this problem, two adaptive operator selection (AOS) mechanisms are introduced to adaptively select the most suitable type of direction information for the specific mutation strategy during the evolutionary process. The new method is named as adaptive direction information based NDi-DE (aNDi-DE). In this way, the good balance between exploration and exploitation can be dynamically achieved. To evaluate the effectiveness of aNDi-DE, the proposed method is applied to the well-known DE/rand/1 algorithm. Through the experimental study, we show that aNDi-DE can effectively improve the efficiency and robustness of NDi-DE. }} @InProceedings{Lotif:2014:CEC, title = {Visualizing the Population of Meta-Heuristics During the Optimization Process Using Self-Organizing Maps}, author = {Marcelo Lotif}, pages = {313--319}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Numerical optimisation}, abstract = { This study proposes a novel Visual Data Mining technique based on Self-Organising Maps (SOM) to visualise the population points of metaheuristic algorithms while they execute their search process. The SOM is used to divide the search space of the optimisation function into bi-dimensional regions, allowing one to perform a visual analysis by mapping the points into the 2-dimensional space, in order to compare various executions of the functions performed with different parameter configurations. The use of these maps as a Visual Data Mining tool aims to visually process the resulting data and identify behavioural patterns of the metaheuristic instances. }} @InProceedings{Lin:2014:CEC, title = {Self-Adaptive Morphable Model Based Multi-View Non-Cooperative {3D} Face Reconstruction}, author = {Kuicheng Lin and Xue Wang and Xuanping Li and Yuqi Tan}, pages = {320--325}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Self-adaptation in evolutionary computation, Multiobjective optimisation}, abstract = { Non-cooperative 3D face reconstruction is very significant in the area of intelligent security. According to non-cooperative 3D face reconstruction, the non-complete information fusion of multi-view face images can be realised to get a more complete face. This paper proposes a non-cooperative 3D face reconstruction method. A multimedia sensor network is employed to detect a person and get face images from different views. View-based active appearance models (View-based AAM) then helps to extract feature points and estimate probable pose angle. A new self-adaptive 3D morphable model based multi-view face geometry reconstruction method is designed to generate a 3D face model with particle swarm optimisation (PSO). As the initial pose estimation is not accurate, particle swarm optimisation is also used to regulate pose estimation results for optimising 3D reconstruction result. "Mirror" strategy is employed to define the invisible part of the face based on the mirror image of the visible part for texture mapping. Experiments have shown that the proposed method can achieve the non-cooperative 3D reconstruction efficaciously. }} @InProceedings{Turky:2014:CEC, title = {Using Electromagnetic Algorithm for Tuning the Structure and Parameters of Neural Networks}, author = {Ayad Turky and Salwani Abdullah}, pages = {326--331}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics}, abstract = { Electromagnetic algorithm is a population based meta-heuristic which imitates the attraction and repulsion of sample points. In this paper, we propose an electromagnetic algorithm to simultaneously tune the structure and parameter of the feed forward neural network. Each solution in the electromagnetic algorithm contains both the design structure and the parameters values of the neural network. This solution later will be used by the neural network to represents its configuration. The classification accuracy returned by the neural network represents the quality of the solution. The performance of the proposed method is verified by using the well-known classification benchmarks and compared against the latest methodologies in the literature. Empirical results demonstrate that the proposed algorithm is able to obtain competitive results, when compared to the best-known results in the literature. }} @InProceedings{Li:2014:CEC, title = {Feature Selection Based on Manifold-Learning with Dynamic Constraint-Handling Differential Evolution}, author = {Zhihui Li and Zhigang Shang and J. J. Liang and B. Y. Qu}, pages = {332--337}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Feature Selection in high dimensional feature space is the main challenge in statistic learning field. In this paper, a novel feature selection method based on manifold learning is proposed. The distance metric weight vector are optimised to maximise the multi-class margin in the manifold embedded in low dimension space, as well as minimise its L1-norm. This multi objectives optimisation problem is solved by a Differential Evolution (DE) with dynamic constraint -handling mechanism. And a criterion to determine the best feature subset based on the optimal weight vector is given. The test result for selecting the optimal feature subset of UCI breast tissue dataset indicates that this real coded feature selection method could find some feature subset which has good classification robustness. }} @InProceedings{Viegas:2014:CEC, title = {Metaheuristics for the {3D} Bin Packing Problem in the Steel Industry}, author = {Joaquim Viegas and Susana Vieira and Joao Sousa and Elsa Henriques}, pages = {338--343}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Discrete and combinatorial optimisation}, abstract = { This work presents heuristic and metaheuristic approaches for addressing the real-world steel cutting problem of a retail steel distributor as a cutting and packing problem. It consists on the cutting of large steel blocks in order to obtain smaller pieces ordered by clients. The problem was formulated as a 3-dimensional residual bin packing problem for minimisation of scrap generation, with guillotine cutting constraint and chips scrap generation. A tabu search and best-fit decreasing (BFD) approaches are proposed and their performance compared to an heuristic and ant colony optimisation (ACO) algorithms. It's shown that the tabu search and best-fit decreasing algorithm are able to reduce the generated scrap by up to 52\% in comparison with the heuristic in [1]. The orders to suppliers were also reduced by up to 35\%. The analysis of the results of the different approaches provide insight onto the most important factors in the problem's scrap minimisation. }} @InProceedings{Gonzalez-Pardo:2014:CEC, title = {A New {CSP} Graph-Based Representation to Resource-Constrained Project Scheduling Problem}, author = {Antonio Gonzalez-Pardo and David Camacho}, pages = {344--351}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation}, abstract = { Resource-Constrained Project Scheduling Problem (RCPSP) is a NP-hard combinatorial problem that consists in scheduling different activities in such a way the resource, precedence, and temporal constraints are satisfied. The main problem when dealing with NP-hard problems is the exponential growth of the computational resources needed to solve the problems. This work is an extension of a previous one, where a new CSP graph-based representation to solve Constraint Satisfaction Problems (CSP) by using Ant Colony Optimisation (ACO) were proposed. This paper studies the behaviour of the CSP graph-based representation when it is applied to a real-world complex problem, in this case the RCPSP. The dataset used in this work has been extracted from Project Scheduling Problem Library (PSPLIB). Experimental results show that the proposed approach provides excellent results, closer to the optimum values published in the PSPLIB repository. Also, it has been analysed how the number of jobs and the number of different execution modes affect the performance of the algorithm. }} @InProceedings{Liu:2014:CECa, title = {Optimization Algorithm for Rectangle Packing Problem Based on Varied-Factor Genetic Algorithm and Lowest Front-Line Strategy}, author = {Haiming Liu and Jiong Zhou and Xinsheng Wu and Peng Yuan}, pages = {352--357}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Heuristics, metaheuristics and hyper-heuristics, Intelligent systems applications}, abstract = { Rectangle packing problem exists widely in manufacturing processes of modern industry, such as cutting of wood, leather, metal and paper, etc. It is also known as a typical NP-Complete combinatorial optimisation problem with geometric nature, which contains two sub-problems, parking problem and sequencing problem of rectangles. Considering the features of the problem, this paper proposes an optimisation algorithm based on an improved genetic algorithm (GA), combined with a lowest front-line strategy for parking rectangles on the sheet. The genetic algorithm is introduced to determine packing sequence of rectangles. To avoid premature convergence or falling into local optima, the traditional GA is improved by changing genetic factors according to quality of solutions obtained during evolution. Numerical experiments were conducted to take an evaluation for the proposed algorithm, along with a comparison with another algorithm. The simulation results show that the proposed algorithm has better performance and can improve use of materials. }} @InProceedings{Farzan:2014:CEC, title = {A Parallel Evolutionary Solution for the Inverse Kinematics of Generic Robotic Manipulators}, author = {Siavash Farzan and Guilherme DeSouza}, pages = {358--365}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Robotics, Evolutionary programming, Parallel and distributed algorithms}, abstract = { This paper is an improvement of our previous work [1]. It provides a robust, fast and accurate solution for the inverse kinematics problem of generic serial manipulators, i.e. any number and any combination of revolute and prismatic joints. Here, we propose further enhancements by applying an evolutionary approach on the previous architecture and explore the effects of different parameters on the performance of the algorithm. The algorithm only requires the Denavit-Hartenberg (D-H) representation of the robot as input and no training or robot-dependent optimisation function is needed. In order to handle singularities and to overcome the possibility of multiple paths in redundant robots, our approach relies on the computation of multiple (parallel) numerical estimations of the inverse Jacobian while it selects the current best path to the desired configuration of the end-effector using an evolutionary algorithm. But unlike other iterative methods, our method achieves submillimeter accuracy in 20 iterations in average. The algorithm was implemented in C/C++ using POSIX threads, and it can be easily expanded to use more threads and/or many-core GPUs. We demonstrate the high accuracy and real-time performance of our method by testing it with five different robots including a 7-DoF redundant robot. Results show that the evolutionary implementation of the algorithm is able to reduce the number of iterations compared to the previous method significantly, while also finding the solution within the specified margin of error. }} @InProceedings{Yue:2014:CEC, title = {Feature Extraction Based on Trimmed Complex Network Representation for Metabolomic Data Classification}, author = {Chen Yue and Zhu Zexuan and Ji Zhen}, pages = {366--370}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation, Computational Intelligence in Bioinformatics, Biometrics, bioinformatics and biomedical applications}, abstract = { Over the last few decades, metabolomics has been widely used to reveal the linkages between metabolite signal levels and physiological states. Metabolomic data are naturally high dimensional and noisy, which poses computational challenges for data analysis. In this study, a novel feature extraction method based on trimmed complex network representation is proposed for metabolomic data classification. Particularly, the proposed method begins with feature selection on the original data, and then a complex network of the selected features is constructed to represent each data sample. After, the network edges are trimmed and a few topological network metrics are extracted as new features for the classification of the samples. The experimental results on a real-world metabolomic data of clinical liver transplantation demonstrate the efficiency of the proposed feature extraction method. }} @InProceedings{Tamura:2014:CEC, title = {Primary Study on Feedback Controlled Differential Evolution}, author = {Kenichi Tamura and Keiichiro Yasuda}, pages = {371--378}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Heuristics, metaheuristics and hyper-heuristics, Differential Evolution: Past, Present and Future}, abstract = { The primary study on feedback controlled Differential Evolution (FCDE) is presented. FCDE is a novel framework of DE with an automatic parameter adjustment mechanism, which controls its search situation (evaluation index) to be a promising situation (reference index) by the error feedback. Its adjustment mechanism consists of three parts: Estimator, Referencer, and Controller. Estimator calculates an evaluation index which quantitatively measures the search situation about the population diversity. Referencer generates a reference index being the ideal target of the evaluation index. Controller operates the DE parameters every generation to make the evaluation index follow the reference index. Further, this paper actually a FCDE method using a typical DE by designing the three parts. The effectiveness of the proposed method is confirmed through computational experiment from viewpoint of the controllability and performance. }} @InProceedings{Yu:2014:CECa, title = {A Route Planning Strategy for the Automatic Garment Cutter Based on Genetic Algorithm}, author = {Wenchao Yu and Linji Lu}, pages = {379--386}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Discrete and combinatorial optimisation, Engineering applications}, abstract = { This paper proposes a route planning algorithm for the automatic garment cutter, a machine extensively used in the clothing industry, aiming at reducing the length and improving the smoothness of quick moving route for the cutter. With proper constraints for the cloth segments and knife-down points, the route planning problem is resolved into a generalised travelling salesman problem (GTSP) of the first category, for which an enhanced genetic algorithm is proposed. In this paper, we firstly outline the procedure of the algorithm and discuss some important details, including individual fitness calculation based on the multistage graph problem, a local search algorithm with 2-opt method, etc. Then a position-reservation crossover operator based on dual-relevancy, and an adaptive mutation operator based on population dispersion are proposed, which can accelerate convergence of the algorithm as well as prevent locking into local minima as much as possible. Finally, experimental tests are performed on the GTSP Instances Library and the data of garment CAD files, which demonstrates the effectiveness of our route planning strategy in terms of both solution quality and running time. }} % Special Session: MoE2-1 Evolutionary Multi-Objective Optimization and Decision Making @InProceedings{Lopez-Herrejon:2014:CEC, title = {Comparative Analysis of Classical Multi-Objective Evolutionary Algorithms and Seeding Strategies for Pairwise Testing of Software Product Lines}, author = {Roberto Erick Lopez-Herrejon and Javier Ferrer and Francisco Chicano and Alexander Egyed and Enrique Alba}, pages = {387--396}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Multiobjective optimisation, Real-world applications}, abstract = { Software Product Lines (SPLs) are families of related software products, each with its own set of feature combinations. Their commonly large number of products poses a unique set of challenges for software testing as it might not be technologically or economically feasible to test of all them individually. SPL pairwise testing aims at selecting a set of products to test such that all possible combinations of two features are covered by at least one selected product. Most approaches for SPL pairwise testing have focused on achieving full coverage of all pairwise feature combinations with the minimum number of products to test. Though useful in many contexts, this single-objective perspective does not reflect the prevailing scenario where software engineers face trade-offs between maximising coverage or minimising the number of products to test. In contrast, our work is the first to propose a classical multi-objective formalisation where both objectives are equally important. We study the application to SPL pairwise testing of four classical multi-objective evolutionary algorithms. We developed three seeding strategies techniques that leverage problem domain knowledge and measured their performance impact on a large and diverse corpus of case studies using two well-known multiobjective quality metrics. Our study identifies performance differences among the algorithms and corroborates that the more domain knowledge leveraged the better the search results. Our findings enable software engineers to select not just one solution (like single-objective techniques) but instead to select from an array of test suite possibilities the one that best matches the economical and technological constraints of their testing context. }} @InProceedings{Li:2014:CECa, title = {An {MOEA/D} with Multiple Differential Evolution Mutation Operators}, author = {Yang Li and Aimin Zhou and Guixu Zhang}, pages = {397--404}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { In evolutionary algorithms, the reproduction operators play an important role. It is arguable that different operators may be suitable for different kinds of problems. Therefore, it is natural to combine multiple operators to achieve better performance. To demonstrate this idea, in this paper, we propose an MOEA/D with multiple differential evolution mutation operators called MOEA/D-MO. MOEA/D aims to decompose a multiobjective optimisation problem (MOP) into a number of single objective optimisation problems (SOPs) and optimise those SOPs simultaneously. In MOEA/D-MO, we combine multiple operators to do reproduction. Three mutation strategies with randomly selected parameters from a parameter pool are used to generate new trial solutions. The proposed algorithm is applied to a set of test instances with different complexities and characteristics. Experimental results show that the proposed combining method is promising. }} @InProceedings{Brands:2014:CEC, title = {Multi-Objective Transportation Network Design: Accelerating Search by Applying {e-NSGAII}}, author = {Ties Brands and Luc Wismans and Eric van Berkum}, pages = {405--412}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making, Multi-objective evolutionary algorithms, Real-world applications}, abstract = { The optimisation of infrastructure planning in a multimodal passenger transportation network is formulated as a multi-objective network design problem, with accessibility, use of urban space by parking, operating deficit and climate impact as objectives. Decision variables are the location of park and ride facilities, train stations and the frequency of public transport lines. For a real life case study the Pareto set is estimated by the Epsilon Non-dominated Sorting Genetic Algorithm (e-NSGAII), since due to high computation time a high performance within a limited number of evaluated solutions is desired. As a benchmark, the NSGAII is used. In this paper Pareto sets from runs of both algorithms are analysed and compared. The results show that after a reasonable computation time, e-NSGAII outperforms NSGAII for the most important indicators, especially in the early stages of algorithm executions. }} @InProceedings{Acampora:2014:CEC, title = {A Comparison of Multi-Objective Evolutionary Algorithms for the Ontology Meta-Matching Problem}, author = {Giovanni Acampora and Hisao Ishibuchi and Autilia Vitiello}, pages = {413--420}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { In recent years, several ontology-based systems have been developed for data integration purposes. The principal task of these systems is to accomplish an ontology alignment process capable of matching two ontologies used for modelling heterogeneous data sources. Unfortunately, in order to perform an efficient ontology alignment, it is necessary to address a nested issue known as ontology meta-matching problem consisting in appropriately setting some regulating parameters. Over years, evolutionary algorithms are appeared to be the most suitable methodology to address this problem. However, almost all of existing approaches work with a single function to be optimised even though a possible solution for the ontology meta-matching problem can be viewed as a compromise among different objectives. Therefore, approaches based on multi objective optimisation are emerging as techniques more efficient than conventional evolutionary algorithms in solving the meta-matching problem. The aim of this paper is to perform a systematic comparison among well-known multi-objective Evolutionary Algorithms (EAs) in solving the meta-matching problem. As shown through computational experiments, among the compared multi-objective EAs, OMOPSO statistically provides the best performance in terms of the well-known measures such as hypervolume, index and coverage of two sets. }} @InProceedings{Mohammadi:2014:CEC, title = {Integrating User Preferences and Decomposition Methods for Many-Objective Optimization}, author = {Asad Mohammadi and Mohammad Nabi Omidvar and Xiaodong Li and Kalyanmoy Deb}, pages = {421--428}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { Evolutionary algorithms that rely on dominance ranking often suffer from a low selection pressure problem when dealing with many-objective problems. Decomposition and user preference based methods can help to alleviate this problem to a great extent. In this paper, a user-preference based evolutionary multi-objective algorithm is proposed that uses decomposition methods for solving many-objective problems. Decomposition techniques that are widely used in multi-objective evolutionary optimisation require a set of evenly distributed weight vectors to generate a diverse set of solutions on the Pareto-optimal front. The newly proposed algorithm, R-MEAD2, improves the scalability of its previous version, R-MEAD, which uses a simplex lattice design method for generating weight vectors. This makes the population size is dependent on the dimension size of the objective space. R-MEAD2 uses a uniform random number generator to remove the coupling between dimension and the population size. This paper shows that a uniform random number generator is simple and able to generate evenly distributed points in a high dimensional space. Our comparative study shows that R-MEAD2 outperforms the dominance-based method R-NSGA-II on many-objective problems. }} @InProceedings{Zapotecas-Martinez:2014:CEC, title = {A Multi-Objective Evolutionary Algorithm Based on Decomposition for Constrained Multi-Objective Optimization}, author = {Saul Zapotecas Martinez and Carlos A. Coello Coello}, pages = {429--436}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making, Multiobjective optimisation, Constraint handling}, abstract = { In spite of the popularity of the Multi-objective Evolutionary Algorithm based on Decomposition (MOEA/D), its use in Constrained Multi-objective Optimisation Problems (Comps) has not been fully explored. In the last few years, there have been a few proposals to extend MOEA/D to the solution of CMOPs. However, most of these proposals have adopted selection mechanisms based on penalty functions. In this paper, we present a novel selection mechanism based on the well-known epsilon-constraint method. The proposed approach uses information related to the neighbourhood adopted in MOEA/D in order to obtain solutions which minimise the objective functions within the allowed feasible region. Our preliminary results indicate that our approach is highly competitive with respect to a state-of-the-art MOEA which solves in an efficient way the constrained test problems adopted in our comparative study. }} % Special Session: MoE2-2 Differential Evolution: Past, Present and Future @InProceedings{Georgieva:2014:CEC, title = {Cooperative {DynDE} for Temporal Data Clustering}, author = {Kristina S. Georgieva and Andries P. Engelbrecht}, pages = {437--444}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future}, abstract = { Temporal data is common in real-world datasets. Clustering of such data allows for relationships between data patterns over time to be discovered. Differential evolution (DE) algorithms have previously been used to cluster temporal data. This paper proposes the cooperative data clustering dynamic DE algorithm (CDCDynDE), which is an adaptation to the data clustering dynamic DE (DCDynDE) algorithm where each population searches for a single cluster centroid. The paper applies the proposed algorithm to a variety of temporal datasets with different frequencies of change, severity of change, dataset dimensions and data migration types. The clustering results of the cooperative data clustering DynDE are compared against the original data clustering DynDE, the re-initialising data clustering DE and the standard data clustering DE. A statistical analysis of these results shows that the cooperative data clustering DynDE algorithm obtains better data clustering solutions to the other three algorithms despite changes in frequency, severity, dimension and data migration types. }} @InProceedings{Liang:2014:CEC, title = {Multi-Objective Differential Evolution Algorithm Based on Fast Sorting and a Novel Constraints Handling Technique}, author = {J. J. Liang and B. Zheng and B. Y. Qu and H. Song}, pages = {445--450}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future, Differential evolution, Multi-objective evolutionary algorithms}, abstract = { In this paper, an improved multi-objective differential evolution algorithm is proposed to solve constraints in multi-objective optimisation. Research has shown that the information of infeasible solutions is also important and can help the algorithm improve the convergence and diversity of solutions. A novel constraint handling method is introduced to ensure that a certain number of good infeasible solutions will be kept in the procedure of evolution to guide the search of the individuals. The proposed method is compared with two other constrained multi-objective differential evolution algorithms and the results show that the proposed method is competitive. }} @InProceedings{Aalto:2014:CEC, title = {A Mutation and Crossover Adaptation Mechanism for Differential Evolution Algorithm}, author = {Johanna Aalto and Jouni Lampinen}, pages = {451--458}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future}, abstract = { A new adaptive Differential Evolution algorithm called EWMA-DECrF is proposed. In original Differential Evolution algorithm three different control parameter values must be pre-specified by the user a priori; Population size, crossover and mutation scale factor. Choosing good parameters can be very difficult for the user, especially for the practitioners. In the proposed algorithm the mutation scale factor and crossover factor is adapted using a mechanism based on exponential weighting moving average, while the population size is kept fixed as in standard Differential Evolution. The algorithm was evaluated by using the set of 25 benchmark functions provided by CEC2005 special session on real-parameter optimisation. It was compared to standard DE/rand/1/bin version and the two other algorithms also based on exponential weighting moving average; EWMA-DE and EWMA-DECr. Results show that proposed algorithm EWMA-DECrF outperformed the other algorithms by its average ranking based on normalised success performance. }} @InProceedings{Segura:2014:CEC, title = {An Analysis of the Automatic Adaptation of the Crossover Rate in Differential Evolution}, author = {Carlos Segura and Carlos A. Coello Coello and Eduardo Segredo and Coromoto Leon}, pages = {459--466}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future}, abstract = { Differential Evolution (DE) is a very efficient metaheuristic for optimisation over continuous spaces which has gained much popularity in recent years. Several parameter control strategies have been proposed to automatically adapt its internal parameters. The most advanced DE variants take into account the feedback obtained in the optimisation process to guide the dynamic setting of the DE parameters. Indeed, the automatic adaptation of the crossover rate (CR) has attracted a lot of research in the last decades. In most of such strategies, the quality of using a given CR value is measured by considering the probability of performing a replacement in the DE selection stage when such a value is applied. One of the main contributions of this paper is to experimentally show that the probability of replacement induced by the application of a given CR value and the quality of the obtained results are not as correlated as expected. This might cause a performance deterioration that avoids the achievement of good quality solutions even in the long-term. In addition, the experimental evaluation developed with a set of optimisation problems of varying complexities clarifies some of the advantages and drawbacks of the different tested strategies. The only component varied among the different tested schemes has been the CR control strategy. The study presented in this paper provides advances in the understanding of the inner working of several state-of-the-art adaptive DE variants. }} @InProceedings{Qin:2014:CEC, title = {Self-Adaptive Differential Evolution with Local Search Chains for Real-Parameter Single-Objective Optimization}, author = {A. K. Qin and Ke Tang and Hong Pan and Siyu Xia}, pages = {467--474}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future}, abstract = { Differential evolution (DE), as a very powerful population-based stochastic optimiser, is one of the most active research topics in the field of evolutionary computation. Self-adaptive differential evolution (SaDE) is a well known DE variant, which aims to relieve the practical difficulty faced by DE in selecting among many candidates the most effective search strategy and its associated parameters. SaDE operates with multiple candidate strategies and gradually adapts the employed strategy and its accompanying parameter setting via learning the preceding behaviour of already applied strategies and their associated parameter settings. Although highly effective, SaDE concentrates more on exploration than exploitation. To enhance SaDE's exploitation capability while maintaining its exploration power, we incorporate local search chains into SaDE following two different paradigms (Lamarckian and Baldwinian) that differ in the ways of using local search results in SaDE. Our experiments are conducted on the CEC-2014 real-parameter single-objective optimisation testbed. The statistical comparison results demonstrate that SaDE with Baldwinian local search chains, armed with suitable parameter settings, can significantly outperform original SaDE as well as classic DE at any tested problem dimensionality. }} @InProceedings{Amin:2014:CEC, title = {Trading-Off Simulation Fidelity and Optimization Accuracy in Air-Traffic Experiments using Differential Evolution}, author = {Rubai Amin and Jiangjun Tang and Mohamed Ellejmi and Stephen Kirby and Hussein A. Abbass}, pages = {475--482}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary simulation-based optimisation}, abstract = { Black-box optimisation relies in many engineering applications on the use of a simulation to obtain a numeric evaluation or a score for a proposed solution. In these cases, the cost of optimisation is mostly a reflection of the cost of running this simulation environment. On the one hand, the higher fidelity the simulation environment is, the longer it is likely to take to evaluate a single solution. Consequently, less solutions are allowed to be evaluated given a time constraint on the running time of the optimisation algorithm. On the other hand, the less fidelity the simulation environment is, the more likely more solutions could be evaluated within the same time constraint. However, the relationship between fidelity and the quality of the final solution obtained by the optimisation is largely unexplored area of research. In this paper, we present an approach for adjusting task load of Air traffic controllers (ATC) in real time by using three different shadow simulators of increasing fidelity and Differential Evolution (DE) as the evolutionary optimisation algorithm. According to air traffic conditions, DE optimises a goal programming model to steer the taskload up or down towards a predefined task load target by generating two ATC requests every 10 minutes. The experiment results suggest demonstrates how a high fidelity simulator can help DE to achieve better results in the absence of any time constraint on running the experiments. However, when there is a tight time constraint is imposed, low fidelity simulators allow DE to explore more solutions in the search space by cutting down on the extra time needed when high fidelity simulators are used. }} % Special Session: MoE2-3 Evolutionary Computation in Combinatorial Optimization @InProceedings{Bennett:2014:CEC, title = {A Hybrid Discrete Particle Swarm Optimisation Method for Grid Computation Scheduling}, author = {Stephen Bennett and Su Nguyen and Mengjie Zhang}, pages = {483--490}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, PSO, Evolutionary Computation for Planning and Scheduling, Evolutionary Computation in Combinatorial Optimisation, Heuristics, metaheuristics and hyper-heuristics}, abstract = { Allocating jobs to heterogeneous machines in grid systems is an important task in computational grid to effectively utilise computational resources. Particle swarm optimisation (PSO) has been recently applied to grid computation scheduling (GCS) problems and shown very promising results as compared to other meta-heuristics in the literature. However, PSO with the traditional position updating mechanism still has problems coping with the discrete nature of GCS. This paper proposed a new updating mechanism for discrete PSO that directly utilise discrete solutions from personal and global best particles. A new local search heuristic has also been proposed to refine solutions found by PSO. The results show that the hybrid PSO is more effective than other existing PSO methods in the literature when tested on two benchmark datasets. The hybrid method is also very efficient, which makes it suitable to deal with large-scale problem instances. }} @InProceedings{Cui:2014:CEC, title = {A Combinatorial Algorithm for the Cardinality Constrained Portfolio Optimization Problem}, author = {Tianxiang Cui and Shi Cheng and Ruibin Bai}, pages = {491--498}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Evolutionary Computation in Combinatorial Optimisation, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Portfolio optimisation is an important problem based on the modern portfolio theory (MPT) in the finance field. The idea is to maximise the portfolio expected return as well as minimising portfolio risk at the same time. In this work, we propose a combinatorial algorithm for the portfolio optimisation problem with the cardinality and bounding constraints. The proposed algorithm hybridises a metaheuristic approach (particle swarm optimisation, PSO) and a mathematical programming method where PSO is used to deal with the cardinality constraints and the math programming method is used to deal with the rest of the model. Computational results are given for the benchmark datasets from the OR-library and they indicate that it is a useful strategy for this problem. We also present the solutions obtained by the CPLEX mixed integer program solver for these instances and they can be used as the criteria for the comparison of algorithms for the same problem in the future. }} @InProceedings{Sabar:2014:CEC, title = {Using Harmony Search with Multiple Pitch Adjustment Operators for the Portfolio Selection Problem}, author = {Nasser R. Sabar and Graham Kendall}, pages = {499--503}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Combinatorial Optimisation, Heuristics, metaheuristics and hyper-heuristics}, abstract = { Portfolio selection is an important problem in the financial markets that seeks to distribute an amount of money over a set of assets where the goal is to simultaneously maximise the return and minimise the risk. In this work, we propose a harmony search algorithm (HSA) for this problem. HSA is a population based algorithm that mimics the musician improvisation process in solving optimisation problems. At each iteration, HSA generates a new solution using a memory procedure which considers all existing solutions and then perturbs them using a pitch adjustment operator. To deal with different instances, and also changes in the problem landscape, we propose an improved HSA that uses multiple pitch adjustment operators. The rationale behind this is that different operators are appropriate for different stages of the search and using multiple operators can enhance the effectiveness of HSA. To evaluate and validate the effectiveness of the proposed HSA, computational experiments are carried out using portfolio selection benchmark instances from the scientific literature. The results demonstrate that the proposed HSA is capable of producing high quality solutions for most of the tested instances when compared with state of the art methods. }} @InProceedings{Smullen:2014:CEC, title = {Genetic Algorithm with Self-Adaptive Mutation Controlled by Chromosome Similarity}, author = {Daniel Smullen and Jonathan Gillett and Joseph Heron and Shahryar Rahnamayan}, pages = {504--511}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Combinatorial Optimisation, Self-adaptation in evolutionary computation, Genetic algorithms}, abstract = { This paper proposes a novel algorithm for solving combinatorial optimisation problems using genetic algorithms (GA) with self-adaptive mutation. We selected the N-Queens problem (N between 8 and 32) as our benchmarking test suite, as they are highly multi-modal with huge numbers of global optima. Optimal static mutation probabilities for the traditional GA approach are determined for each N to use as a best-case scenario benchmark in our conducted comparative analysis. Despite an unfair advantage with traditional GA using optimised fixed mutation probabilities, in large problem sizes (where N $>$ 15) multi-objective analysis showed the self adaptive approach yielded a 100 to 584 percents improvement in the number of distinct solutions generated; the self-adaptive approach also produced the first distinct solution faster than traditional GA with a 1.90 to 70.0 percents speed improvement. Self-adaptive mutation control is valuable because it adjusts the mutation rate based on the problem characteristics and search process stages accordingly. This is not achievable with an optimal constant mutation probability which remains unchanged during the search process. }} @InProceedings{Yu:2014:CECb, title = {Chemical Reaction Optimization for the Set Covering Problem}, author = {James J.Q. Yu and Albert Y.S. Lam and Victor O.K. Li}, pages = {512--519}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial optimisation, Evolutionary Computation in Combinatorial Optimisation}, abstract = { The set covering problem (SCP) is one of the representative combinatorial optimisation problems, having many practical applications. This paper investigates the development of an algorithm to solve SCP by employing chemical reaction optimisation (CRO), a general-purpose metaheuristic. It is tested on a wide range of benchmark instances of SCP. The simulation results indicate that this algorithm gives outstanding performance compared with other heuristics and metaheuristics in solving SCP. }} @InProceedings{Sabar:2014:CECa, title = {Aircraft Landing Problem Using Hybrid Differential Evolution and Simple Descent Algorithm}, author = {Nasser R. Sabar and Graham Kendall}, pages = {520--527}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Combinatorial Optimisation, Real-world applications}, abstract = { The aircraft landing problem (ALP) is a practical and challenging optimisation problem for the air traffic industry. ALP involves allocating a set of aircraft to airport runways and allocating landing times for which the goal is to minimise the total cost of landing deviation from the preferred target times. Differential evolution (DE) is a population based algorithm that has been shown to be an effective algorithm for solving continuous optimisation problems. However, DE can suffer from slow convergence when used for combinatorial optimisation problems, thus hindering its ability to return good quality solutions in these domains. To address this we propose a hybrid algorithm that combines differential evolution with a simple descent algorithm. DE is responsible for exploring new regions in the search space, whilst the descent algorithm focuses the search around the area currently being explored. Experimenting with widely used ALP benchmark instances, we demonstrate that the proposed hybrid algorithm performs better than DE without the simple descent algorithm. Furthermore, performance comparisons with other algorithms from the scientific literature demonstrate that our hybrid algorithm performs better, or at least comparably, in terms of both solution quality and computational time. }} % Special Session: MoE2-4 Artificial Bee Colony Algorithms and their Applications @InProceedings{Li:2014:CECb, title = {Search-Evasion Path Planning for Submarines Using the Artificial Bee Colony Algorithm}, author = {Bai Li and Raymond Chiong and Ligang Gong}, pages = {528--535}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications, Single Objective Numerical Optimisation, Robotics}, abstract = { Submarine search-evasion path planning aims to acquire an evading route for a submarine so as to avoid the detection of hostile anti-submarine searchers such as helicopters, aircraft and surface ships. In this paper, we propose a numerical optimisation model of search-evasion path planning for invading submarines. We use the Artificial Bee Colony (ABC) algorithm, which has been confirmed to be competitive compared to many other nature-inspired algorithms, to solve this numerical optimisation problem. In this work, several search-evasion cases in the two-dimensional plane have been carefully studied, in which the anti-submarine vehicles are equipped with sensors with circular footprints that allow them to detect invading submarines within certain radii. An invading submarine is assumed to be able to acquire the real-time locations of all the anti-submarine searchers in the combat field. Our simulation results show the efficacy of our proposed dynamic route optimisation model for the submarine search-evasion path planning mission. }} @InProceedings{Fatnassi:2014:CEC, title = {A Bee Colony Algorithm for Routing Guided Automated Battery-Operated Electric Vehicles in Personal Rapid Transit Systems}, author = {Ezzeddine Fatnassi and Olfa Chebbi and Jouhaina Chaouachi}, pages = {536--543}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications}, abstract = { A personal rapid transit (PRT) system is an on demand transportation service that uses guided automated vehicles. This paper introduces an artificial bee colony (ABC) heuristic for solving the routing problem associated with PRTs. An ABC is a swarm-based heuristic that mimics the behaviour of bees. An enhanced version of this algorithm, in which we add a specific method to escape from local optima, is presented in this paper. Experimental results for 1320 randomly generated instances are also presented and analysed. }} @InProceedings{Fong:2014:CEC, title = {A Novel Hybrid Approach for Curriculum Based Course Timetabling Problem}, author = {Cheng Weng Fong and Hishammuddin Asmuni and Way Shen Lam and Barry McCollum and Paul McMullan}, pages = {544--550}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications}, abstract = { This work applies a hybrid approach in solving the university curriculum-based course timetabling problem as presented as part of the 2nd International Timetabling Competition 2007 (ITC2007). The core of the hybrid approach is based on an artificial bee colony algorithm. Past methods have applied artificial bee colony algorithms to university timetabling problems with high degrees of success. Nevertheless, there exist inefficiencies in the associated search abilities in term of exploration and exploitation. To improve the search abilities, this work introduces a hybrid approach entitled Nelder-mead great deluge artificial bee colony algorithm (NMGD-ABC) where it combined additional positive elements of particle swarm optimisation and great deluge algorithm. In addition, Nelder-mead local search is incorporated into the great deluge algorithm to further enhance the performance of the resulting method. The proposed method is tested on curriculum-based course timetabling as presented in the ITC2007. Experimental results reveal that the proposed method is capable of producing competitive results as compared with the other approaches described in literature. }} @InProceedings{Bulut:2014:CEC, title = {A Discrete Artificial Bee Colony Algorithm for the Economic Lot Scheduling Problem with Returns}, author = {Onder Bulut and M. Fatih Tasgetiren}, pages = {551--557}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { In this study, we model the Economic Lot Scheduling problem with returns (ELSPR) under the basic period (BP) policy with power-of-two (PoT) multipliers, and solve it with a discrete artificial bee colony (DABC) algorithm. Tang and Teunter [1] is the first to consider the well-known economic lot scheduling problem (ELSP) with return flows and re-manufacturing opportunities. Teunter et al. [2] and Zanoni et al. [3] recently extended this first study by proposing heuristics for the common cycle policy and for a modified basic period policy, respectively. As Zanoni et al. [3], we restrict the study to consider independently managed serviceable inventory to test the performance of the proposed algorithm. Our study, to the best of our knowledge, is the first to solve ELSPR using a meta-heuristic. ABC is a swarm-intelligence-based meta-heuristic inspired by the intelligent foraging behaviours of honeybee swarms. In this study, we implement the ABC algorithm with some modifications to handle the discrete decision variables. In the algorithm, we employ two different constraint handling methods in order to have both feasible and infeasible solutions within the population. Our DABC is also enriched with a variable neighbourhood search (VNS) algorithm to further improve the solutions. We test the performance of our algorithm on the two problem instances used in Zanoni et al. [3]. The numerical study depicts that the proposed algorithm performs well under the BP-PoT policy and it has the potential of improving the best known solutions when we relax BP, PoT and independently managed serviceable inventory restrictions in the future. }} @InProceedings{Liang:2014:CECa, title = {Artificial Bee Colony for Workflow Scheduling}, author = {Yun-Chia Liang and Hsiang-Ling Chen and Yung-Hsiang Nien}, pages = {558--564}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications}, abstract = { Cloud computing is the provision of computing resource services from which users can obtain resources via network to tackle their demands. In recent years, with fast growing information technology, more users apply this service; as a result, the demand has increased dramatically. In addition, most of the complex tasks are represented by work flow and executed in the cloud. Therefore, as service providers face this increasing demand, how to schedule the work flow and reduce the response time becomes a critical issue. This research integrates the concept of project scheduling with the work flow scheduling problem to formulate a mathematical model, which expects to minimise the total completion time. Two Artificial Bee Colony algorithms are proposed to solve the work flow scheduling optimisation problem. The performance of ABC is compared with the optimal solutions obtained by Gurobi optimiser on the instance containing different sizes of work flows. The results show that ABC can be considered a practical method for complicated work flow scheduling problems in the cloud computing environment. }} @InProceedings{Madureira:2014:CEC, title = {Cooperation Mechanism For Distributed Resource Scheduling Through Artificial Bee Colony Based Self-Organized Scheduling System}, author = {Ana Madureira and Bruno Cunha and Ivo Pereira}, pages = {565--572}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial Bee Colony Algorithms and Their Applications, Intelligent systems applications, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { In this paper a Cooperation Mechanism for Distributed Scheduling based on Bees based Computing is proposed. Where multiple self-interested agents can reach agreement over the exchange of operations on cooperative resources. Agents must collaborate to improve their local solutions and the global schedule. The proposed cooperation mechanism is able to analyse the scheduling plan generated by the Resource Agents and refine it by idle times reducing taking advantage from cooperative and the self-organised behaviour of Artificial Bee Colony technique. The computational study allows concluding about statistical evidence that the cooperation mechanism influences significantly the overall system performance. }} @InProceedings{Jana:2014:CEC, title = {Particle Swarm Optimization with Population Adaptation}, author = {Nanda Dulal Jana and Swagatam Das and Jaya Sil}, pages = {573--578}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Heuristics, metaheuristics and hyper-heuristics, Biometrics, bioinformatics and biomedical applications}, abstract = { The Particle Swarm Optimisation (PSO) algorithm is a novel population based swarm algorithm has shown good performance on well-known numerical test problems. However, PSO has tends to suffer from premature convergence on multimodal test problems. This is due to lack of diversity of population in search space and leads to stuck at local optima and ultimately fitness stagnation of the population. To enhance the performance of PSO algorithms, in this paper, we propose a method of population adaptation (PA). The proposed method can identify the moment when the population diversity is poor or the population stagnates by measuring the Euclidean distance between particle position and particles average position of a population. When stagnation in the population is identified, the population will be regenerated by normal distribution to increase diversity in the population. The population adaptation is incorporated into the PSO algorithm and is tested on a set of 13 scalable CEC05 benchmark functions. The results show that the proposed population adaptation algorithm can significantly improve the performance of the PSO algorithm with standard PSO, ATREPSO and ARPSO. }} % Special Session: TuE1-1 Evolutionary Computation for Planning and Scheduling @InProceedings{Liu:2014:CECb, title = {A Benchmark Generator for Dynamic Capacitated Arc Routing Problems}, author = {Min Liu and Hemant Singh and Tapabrata Ray}, pages = {579--586}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Planning and Scheduling, Discrete and combinatorial optimisation, Dynamic and uncertain environments}, abstract = { Capacitated arc routing problems (CARPs) are usually modelled as static problems, where information is known in advance and assumed to remain constant during the course of optimisation. However, in practice, many factors such as demand, road accessibility, vehicle availability etc. change during the course of a mission and the route of each vehicle must be reconfigured dynamically. This problem is referred to as dynamic capacitated arc routing problem (DCARP) and there have been limited attempts to solve such problems in the past. Lack of standard DCARP benchmarks is one of the key factors limiting research in this direction. This paper introduces a benchmark generator for DCARPs considering interruptions/changes that are likely to occur in realistic scenarios. These benchmarks can be used to evaluate the strengths and the weaknesses of various optimisation algorithms attempting to solve realistic DCARP problems. }} @InProceedings{Zheng:2014:CEC, title = {A Co-Evolutionary Teaching-Learning-Based Optimization Algorithm for Stochastic {RCPSP}}, author = {Huan-yu Zheng and Ling Wang and Sheng-yao Wang}, pages = {587--594}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Planning and Scheduling}, abstract = { A co-evolutionary teaching-learning-based optimisation (CTLBO) algorithm is proposed in this paper to solve the stochastic resource-constrained project scheduling problem (SRCPSP). The activity list is used for encoding, and resource-based policies are used for decoding. Also, a new competition phase is developed to select the best solution of each class as the teacher. To make two classes evolve cooperatively, both the teacher phase and student phase of the TLBO are modified. Moreover, Taguchi method of design of experiments is used to investigate the effect of parameter setting. Computational results are provided based on the well-known PSPLIB with certain probability distributions. The comparisons between the CTLBO and some state-of-the-art algorithms are provided. It shows that the CTLBO is more effective in solving the problems with medium to large variance. }} @InProceedings{Liu:2014:CECc, title = {A Memetic Algorithm with a New Split Scheme for Solving Dynamic Capacitated Arc Routing Problems}, author = {Min Liu and Hemant Singh and Tapabrata Ray}, pages = {595--602}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Planning and Scheduling, Discrete and combinatorial optimisation, Dynamic and uncertain environments}, abstract = { Capacitated arc routing problems (CARPs) are usually modelled as static problems, where all information about the problem is known in advance and assumed to remain constant during the course of optimisation. However, in practice, many factors such as demand, road accessibility, vehicle availability etc. change during the course of a mission and the routes of each vehicle must be reconfigured dynamically. This problem is referred to as dynamic capacitated arc routing problem (DCARP). In this study, a memetic algorithm with a new split scheme for DCARPs is proposed. This algorithm is capable to solve DCARPs with variations in vehicle availability, road accessibility, added/cancelled tasks or demands and traffic congestion. The algorithm is also capable of solving static CARPs. The performance of the algorithm is reported on a 10-node and three 100-node examples in order to demonstrate the efficacy of the algorithm in solving static and dynamic problems. }} @InProceedings{Yuan:2014:CEC, title = {Agile Earth Observing Satellites Mission Planning Using Genetic Algorithm Based on High Quality Initial Solutions}, author = {Zang Yuan and Yingwu Chen and Renjie He}, pages = {603--609}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Planning and Scheduling}, abstract = { This paper presents an improved genetic algorithm to solve the agile earth observing satellite mission planning problem. We study how to rapidly generate high quality initial solutions, and four generation strategies are proposed. The effect of the settings of operator parameters on the performance of the algorithm is analysed. The experiment results show that the genetic algorithm based on high quality initial solutions generated by Hybrid Random Heuristic Strategy (HRHS) is more effective in solving the agile satellite mission planning problem, but in a certain time cost. We expect that our results will provide insights for the future application of genetic algorithm to satellites mission planning problems. }} @InProceedings{Tang:2014:CEC, title = {Behavioral Learning of Aircraft Landing Sequencing Using a Society of Probabilistic Finite State Machines}, author = {Jiangjun Tang and Hussein A. Abbass}, pages = {610--617}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Planning and Scheduling}, abstract = { Air Traffic Control (ATC) is a complex safety critical environment. A tower controller would be making many decisions in real-time to sequence aircraft. While some optimisation tools exist to help the controller in some airports, even in these situations, the real sequence of the aircraft adopted by the controller is significantly different from the one proposed by the optimisation algorithm. This is due to the very dynamic nature of the environment. The objective of this paper is to test the hypothesis that one can learn from the sequence adopted by the controller some strategies that can act as heuristics in decision support tools for aircraft sequencing. This aim is tested in this paper by attempting to learn sequences generated from a well-known sequencing method that is being used in the real world. The approach relies on a genetic algorithm (GA) to learn these sequences using a society Probabilistic Finite-state Machines (PFSMs). Each PFSM learns a different sub-space; thus, decomposing the learning problem into a group of agents that need to work together to learn the overall problem. Three sequence metrics (Levenshtein, Hamming and Position distances) are compared as the fitness functions in GA. As the results suggest, it is possible to learn the behaviour of the algorithm/heuristic that generated the original sequence from very limited information. }} @InProceedings{Hunt:2014:CEC, title = {Evolving Machine-Specific Dispatching Rules for a Two-Machine Job Shop using Genetic Programming}, author = {Rachel Hunt and Mark Johnston and Mengjie Zhang}, pages = {618--625}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Evolutionary Computation for Planning and Scheduling}, abstract = { Job Shop Scheduling (JSS) involves determining a schedule for processing jobs on machines to optimise some measure of delivery speed or customer satisfaction. We investigate a genetic programming based hyper-heuristic (GPHH) approach to evolving dispatching rules for a two-machine job shop in both static and dynamic environments. In the static case the proposed GPHH method can represent and discover optimal dispatching rules. In the dynamic case we investigate two representations (using a single rule at both machines and evolving a specialised rule for each machine) and the effect of changing the training problem instances throughout evolution. Results show that relative performance of these methods is dependent on the testing instances. }} % Special Session: TuE1-2 Swarm Intelligence for Real-World Engineering Optimization @InProceedings{Zheng:2014:CECa, title = {An Enhanced Non-Dominated Sorting Based Fruit Fly Optimization Algorithm for Solving Environmental Economic Dispatch Problem}, author = {Xiaolong Zheng and Ling Wang and Shengyao Wang}, pages = {626--633}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation, Multi-objective evolutionary algorithms}, abstract = { A fruit fly optimisation algorithm based on the enhanced non-dominated sorting (ESFOA) is proposed to solve the environmental economic dispatch (EED) problem. To measure the difference between two non-dominated solutions, the concept of the enhanced non-dominance is defined, and the degrees of dominance and non-dominance are presented. To enhance the parallel search ability, multiple fruit flies groups are used to perform evolutionary search in the ESFOA. In the vision-based search process, the best fruit fly is determined according to the enhanced non-dominance value. To guarantee the feasibility of the new solutions, an effective heuristic mechanism to handle constraints is adopted to repair the infeasible solutions. Meanwhile, an external archive is used to store the non-dominated solutions. The influence of parameter setting is investigated based on the Taguchi method of design of experiment, and a suitable parameter setting is suggested. Finally, numerical tests are carried out by using the IEEE 30-bus benchmark. The comparisons to some existing methods by using the technique for order preference by similarity to ideal solution (TOPSIS) demonstrate the effectiveness of the proposed algorithm. }} @InProceedings{Niu:2014:CEC, title = {Particle Swarm Optimization for Integrated Yard Truck Scheduling and Storage Allocation Problem}, author = {Ben Niu and Ting Xie and Qiqi Duan and Lijing Tan}, pages = {634--639}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { The Integrated Yard Truck Scheduling and Storage Allocation Problem (YTS-SAP) is one of the major optimisation problems in container port which minimises the total delay for all containers. To deal with this NP-hard scheduling problem, standard particle swarm optimisation (SPSO) and a local version PSO (LPSO) are developed to obtain the optimal solutions. In addition, a simple and effective 'problem mapping' mechanism is used to convert particle position vector into scheduling solution. To evaluate the performance of the proposed approaches, experiments are conducted on different scale instances to compare the results obtained by GA. The experimental studies show that PSOs outperform GA in terms of computation time and solution quality. }} @InProceedings{Liu:2014:CECd, title = {Similarity- and Reliability-Assisted Fitness Estimation for Particle Swarm Optimization of Expensive Problems}, author = {Tong Liu and Chaoli Sun and Jianchao Zeng and Yaochu Jin}, pages = {640--646}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { As a population-based meta-heuristic technique for global search, particle swarm optimisation (PSO) performs quite well on a variety of problems. However, the requirement on a large number of fitness evaluations poses an obstacle for the PSO algorithm to be applied to solve complex optimisation problems with computationally expensive objective functions. This paper extends a fitness estimation strategy for PSO (FESPSO) based on its search dynamics to reduce fitness evaluations using the real fitness function. In order to further save the fitness evaluations and improve the estimation accuracy, a similarity measure and a reliability measure are introduced into the FESPSO. The similarity measure is used to judge whether the fitness of a particle will be estimated or evaluated using the real fitness function, and the reliability measure is adopted to determine whether the approximated value will be trusted. Experimental results on six commonly used benchmark problems show the effectiveness and competitiveness of our proposed algorithm. Preliminary empirical analysis of the search behaviour is also performed to illustrate the benefit of the proposed estimation mechanism. }} @InProceedings{Niu:2014:CECa, title = {Binary Bacterial Foraging Optimization for Solving 0/1 Knapsack Problem}, author = {Ben Niu and Ying Bi}, pages = {647--652}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Knapsack problem is famous NP-complete problem where one has to maximise the benefit of objects in a knapsack without exceeding its capacity. In this paper, a binary bacterial foraging optimisation (BBFO) is proposed to find solutions of 0/1 knapsack problems. The original BFO chemotaxis equation is modified to operate in discrete space by using a mapping function, where some new variables and parameter, i.e., binary matrix y, logistic transformation S, and limiting transformation L is built to transform the bacterial position to a binary matrix. By using this schema, the proposed BBFO model can also be easily applied in other discrete problem solving. To further validate the efficiency of the BFO-based approach, an improved version BFO named BFO with linear decreasing chemotaxis step (BFO-LDC) is used to evaluate on six different instances. Comparisons with particle swarm optimisation (PSO) and original BFO are presented and discussed. }} @InProceedings{Kizilay:2014:CEC, title = {A Discrete Artificial Bee Colony Algorithm for the Parallel Machine Scheduling Problem in {DYO} Painting Company}, author = {Damla Kizilay and M. Fatih Tasgetiren and Onder Bulut and Bilgehan Bostan}, pages = {653--660}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation, Artificial Bee Colony Algorithms and Their Applications, Heuristics, metaheuristics and hyper-heuristics}, abstract = { This paper presents a discrete artificial bee colony algorithm to solve the assignment and parallel machine scheduling problem in DYO paint company. The aim of this paper is to develop some algorithms to be employed in the DYO paint company by using their real-life data in the future. Currently, in the DYO paint company; there exist three types of filling machines groups. These are automatic, semiautomatic and manual machine groups, where there are several numbers of identical machines. The problem is to first assign the filling production orders (jobs) to machine groups. Then, filling production orders assigned to each machine group should be scheduled on identical parallel machines to minimise the sum of makespan and the total weighted tardiness. We also develop a traditional genetic algorithm to solve the same problem. The computational results show that the DABC algorithm outperforms the GA on set of benchmark problems we have generated. }} @InProceedings{Wang:2014:CECa, title = {Locality-Sensitive Hashing Based Multiobjective Memetic Algorithm for Dynamic Pickup and Delivery Problems}, author = {Fangxiao Wang and Yuan Gao and Zexuan Zhu}, pages = {661--666}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation, Memetic, multi-meme and hybrid algorithms, Multiobjective optimisation}, abstract = { This paper proposes a locality-sensitive hashing based multiobjective memetic algorithm namely LSH-MOMA for solving pickup and delivery problems with dynamic requests (DPDPs for short). Particularly, LSH-MOMA is designed to find the solution route of a DPDP by optimising objectives namely workload and route length in an evolutionary manner. In each generation of LSH-MOMA, locality sensitive hashing based rectification and local search are imposed to repair and refine the individual candidate routes. LSH-MOMA is evaluated on three simulated DPDPs of different scales and the experimental results demonstrate the efficiency of the method. }} % Special Session: TuE1-3 Complex Networks and Evolutionary Computation @InProceedings{Wu:2014:CECa, title = {A Compression Optimization Algorithm for Community Detection}, author = {Jianshe Wu and Lin Yuan and Qingliang Gong and Wenping Ma and Jingjing Ma and Yangyang Li}, pages = {667--671}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation, Hybrid evolutionary Computational Methods for Complex Optimisation Problems, Evolutionary Computation for Grouping and Graph-based Clustering}, abstract = { Community detection is important in understanding the structures and functions of complex networks. Many algorithms have been proposed. The most popular algorithms detect the communities through optimising a criterion function known as modularity, which suffer from the resolution limit problem. Some algorithms require the number of communities as a prior. In this paper, a non-modularity based compression optimisation algorithm for community detection is proposed without any prior knowledge, which is efficient and is suitable for large scale networks. }} @InProceedings{Wang:2014:CECb, title = {Decomposition Based Multiobjective Evolutionary Algorithm for Collaborative Filtering Recommender Systems}, author = {Shanfeng Wang and Maoguo Gong and Lijia Ma and Qing Cai and Licheng Jiao}, pages = {672--679}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation}, abstract = { With the rapid expansion of the information on the Internet, recommender systems play an important role in filtering insignificant information and recommend satisfactory items to users. Accurately predicting the preference of users is the main priority of recommendation. Diversity is also an important objective in recommendation, which is achieved by recommending items from the so-called long tail of goods. Traditional recommendation techniques lay more emphasis on accuracy and overlook diversity. Simultaneously optimising the accuracy and diversity is a multiobjective optimisation problem, in which the two objectives are contradictory. In this paper, A multiobjective evolutionary algorithm based on decomposition is proposed for recommendation, which maximises the predicted score and the popularity of items simultaneously. This algorithm returns lots of non-dominated solutions and each solution is a trade-off between the accuracy and diversity. The experiment shows that our algorithm can provide a series of recommendation results with different precision and diversity to a user. }} @InProceedings{Mu:2014:CEC, title = {A Memetic Algorithm Using Local Structural Information for Detecting Community Structure in Complex Networks}, author = {Caihong Mu and Jin Xie and Ruochen Liu and Licheng Jiao}, pages = {680--686}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation}, abstract = { Community detection has received a great deal of attention in recent years. Modularity is the most used and best known quality function for measuring the quality of a partition of a network. Based on the optimisation of modularity, we proposed a memetic algorithm with a local search operator to detect community structure. The local search operator uses a quality function of local community tightness based on structural similarity. In addition, the tactics of vertex mover is used for reassigning vertices to neighbouring communities to improve the partition result. Experiments on real-world networks and computer-generated networks show the effectiveness of our algorithm. }} @InProceedings{Song:2014:CEC, title = {Ant Colony Clustering Based on Sampling for Community Detection}, author = {Xiangjing Song and Junzhong Ji and Cuicui Yang and Xiuzhen Zhang}, pages = {687--692}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation, Classification, clustering and data analysis, Data mining}, abstract = { Community structure detection in large-scale complex networks has been intensively investigated in recent years. In this paper, we propose a new framework which employs the ant colony clustering algorithm based on sampling to discover communities in large-scale complex networks. The algorithm firstly samples a small number of representative nodes from the large-scale network; secondly it uses the ant colony clustering algorithm to cluster the sampled nodes; thirdly it assigns the un-sampled nodes into the detected communities according to the similarity metric; finally it merges the initial clustering result to sustainably increase the modularity function value of the detection results. A significant advantage of our algorithm is that the sampling method greatly reduces the scale of the problem. Experimental results on computer-generated and real world networks show the efficiency of our method. }} @InProceedings{Kuang:2014:CECa, title = {A Differential Evolution Box-Covering Algorithm for Fractal Dimension on Complex Networks}, author = {Li Kuang and Zhiyong Zhao and Feng Wang and Yuanxiang Li and Fei Yu and Zhijie Li}, pages = {693--699}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation}, abstract = { The fractal property are discovered on complex networks through renormalisation procedure, which is implemented by box-covering method. The unsolved problem of box-covering method is to find the minimum number of boxes to cover the whole network. Here, we introduce a differential evolution box-covering algorithm based on greedy graph colouring approach. We apply our algorithm on some benchmark networks with different structures, such as the E.coli metabolic network, which has low clustering coefficient and high modularity, the Clustered scale-free network, which has high clustering coefficient and low modularity, and some community networks (the Politics books network, the Dolphins network, and the American football games network), which have high clustering coefficient. Experimental results show that our DEBC algorithm can get better results than state of art algorithms in most cases, especially has significant improvement in clustered community networks. }} @InProceedings{Mu:2014:CECa, title = {An Intelligent Ant Colony Optimization for Community Detection in Complex Networks}, author = {Caihong Mu and Jian Zhang and Licheng Jiao}, pages = {700--706}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation}, abstract = { Many systems in social world can be represented by complex networks. It is of great significance to detect the community structure and analyse the functions for networks. In recent years, plenty of research and works have been focused on this problem. In this paper, we propose an enhanced algorithm based on ant colony optimisation (ACO) for the community detection problems. In order to avoid redundant computing in ACO, we divide the ant colony into two groups, original group and intelligent group, which search the solution space simultaneously. In the intelligent group, due to the locus-based adjacency representation of the solution, we let some of them have an ability of self-learning and others can learn from the optimal solutions proactively. Experiments on synthetic and real-life networks show the proposed algorithm can explore in an efficient and stable way. }} % Special Session: TuE1-4 Evolutionary Algorithms with Statistical and Machine Learning Techniques @InProceedings{Zhang:2014:CECa, title = {{HMOEDA\_LLE}: A Hybrid Multi-Objective Estimation of Distribution Algorithm Combining Locally Linear Embedding}, author = {Yuzhen Zhang and Guangming Dai and Lei Peng and Maocai Wang}, pages = {707--714}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques, Multi-objective evolutionary algorithms, Estimation of distribution algorithms}, abstract = { Based on the regularity that: the Pareto set of a continuous m-objectives problem is a piecewise continuous (m-1)-dimensional manifold, a novel hybrid multi-objective optimisation algorithm is proposed in this paper. In the early evolutionary stage, traditional crossover and mutation operations are used to produce offspring, in addition, the locally linear embedding (LLE) with small neighbour parameter approach is introduced to learn the local geometry of the manifold. When certain regularity in population's distribution is detected, new offspring are sampled from the probability models created by the statistical distribution information. An entropy-based criterion is imported to determine the switching time of the two different phases of evolutionary search. The proposed hybrid multi-objective estimation of distribution algorithm combining locally linear embedding (HMOEDA_LLE) adopts several widely used test problems to conduct the comparison experiments with two state-of-the-art multi-objective evolutionary algorithms NSGA-II and RM-MEDA. The simulated results show the effectiveness of the entropy-based criterion and the proposed algorithm has better optimisation performance. }} @InProceedings{Liu:2014:CECe, title = {Behavioral Study of the Surrogate Model-Aware Evolutionary Search Framework}, author = {Bo Liu and Qin Chen and Qingfu Zhang and Goerges Gielen and Vic Grout}, pages = {715--722}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques}, abstract = { The surrogate model-aware evolutionary search (SMAS) framework is an emerging model management method for surrogate model assisted evolutionary algorithms (SAEAs). SAEAs based on SMAS outperform several state-of-the-art SAEAs using other model management methods and show promising results in real-world computationally expensive optimisation problems. However, there is little behavioural study of the SMAS framework, and appropriate rules for its search strategy, training data selection and key parameter selection for different types of problems have not been provided yet. In this paper, with a newly proposed training data selection method, the SMAS framework's behaviour with different search strategies and training data selection methods is investigated. The empirical rules in terms of problem characteristics are obtained and the method to construct an SAEA based on the SMAS framework is updated. Experiments using 24 widely used benchmark test problems and the test problems in the CEC 2014 competition of computationally expensive optimisation are carried out, which validate the proposed empirical rules. }} @InProceedings{Zhang:2014:CECb, title = {A Clustering Based Multiobjective Evolutionary Algorithm}, author = {Hu Zhang and Shenmin Song and Aimin Zhou and Xiao-Zhi Gao}, pages = {723--730}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques}, abstract = { In this paper, we propose a clustering based multiobjective evolutionary algorithm (CLUMOEA) to deal with the multiobjective optimisation problems with irregular Pareto front shapes. CLUMOEA uses a k-means clustering method to discover the population structure by partitioning the solutions into several clusters, and it only allows the solutions in the same cluster to do the reproduction. To reduce the computational cost and balance the exploration and exploitation, the clustering process and evolutionary process are integrated together and they are performed simultaneously. In addition to the clustering, CLUMOEA also uses a distance tournament selection to choose the more similar mating solutions to accelerate the convergence. Besides, a cosine nondominated selection method considering the location and distance information of the solutions are further presented to construct the final population with good diversity. The experimental results show that, compared with some state-of-the-art algorithms, CLUMOEA has significant advantages on dealing with the given test problems with irregular Pareto front shapes. }} @InProceedings{Li:2014:CECc, title = {Creating Stock Trading Rules Using Graph-Based Estimation of Distribution Algorithm}, author = {Xianneng Li and Wen He and Kotaro Hirasawa}, pages = {731--738}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, genetic network programming, Evolutionary Algorithms with Statistical and Machine Learning Techniques, Estimation of distribution algorithms}, abstract = { Though there are numerous approaches developed currently, exploring the practical applications of estimation of distribution algorithm (EDA) has been reported to be one of the most important challenges in this field. This paper is dedicated to extend EDA to solve one of the most active research problems, stock trading, which has been rarely revealed in the EDA literature. A recent proposed graph-based EDA called reinforced probabilistic model building genetic network programming (RPMBGNP) is investigated to create stock trading rules. With its distinguished directed graph-based individual structure and the reinforcement learning-based probabilistic modelling, we demonstrate the effectiveness of RPMBGNP for the stock trading task through real-market stock data, where much higher profits are obtained than traditional non-EDA models. }} @InProceedings{Wong:2014:CEC, title = {Grammar Based Genetic Programming with {Bayesian} Network}, author = {Pak-Kan Wong and Leung-Yau Lo and Man-Leung Wong and Kwong-Sak Leung}, pages = {739--746}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Evolutionary Algorithms with Statistical and Machine Learning Techniques, Estimation of distribution algorithms}, abstract = { Grammar-Based Genetic Programming (GBGP) improves the search performance of Genetic Programming (GP) by formalising constraints and domain specific knowledge in grammar. The building blocks (i.e. the functions and the terminals) in a program can be dependent. Random crossover and mutation destroy the dependence with a high probability, hence breeding a poor program from good programs. Understanding on the syntactic and semantic in the grammar plays an important role to boost the efficiency of GP by reducing the number of poor breeding. Therefore, approaches have been proposed by introducing context sensitive ingredients encoded in probabilistic models. In this paper, we propose Grammar-Based Genetic Programming with Bayesian Network (BGBGP) which learns the dependence by attaching a Bayesian network to each derivation rule and demonstrates its effectiveness in two benchmark problems. }} @InProceedings{Krawczyk:2014:CEC, title = {A First Attempt on Evolutionary Prototype Reduction for Nearest Neighbor One-Class Classification}, author = {Bartosz Krawczyk and Isaac Triguero and Salvador Garcia and Michal Wozniak and Francisco Herrera}, pages = {747--753}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques}, abstract = { Evolutionary prototype reduction techniques are data pre-processing methods originally developed to enhance the nearest neighbour rule. They reduce the training data by selecting or generating representative examples of a given problem. These algorithms have been designed and widely analysed in standard classification providing very competitive results. However, its application scope can be extended to many other specific domains, such as one-class classification, in which its way of working is very interesting in order to reduce computational complexity and sensitivity to noisy data. In this contribution, we perform a first study on the usefulness of evolutionary prototype reduction methods for one-class classification. To do so, we will focus on two recent evolutionary approaches that follow very different strategies: selection and generation of examples from the training data. Both alternatives provide a resulting preprocessed data set that will be used later by a nearest neighbour one-class classifier as its training data. The results achieved support that these data reduction techniques are suitable tools to improve the performance of the nearest neighbour one-class classification. }} % Plenary Poster Session: PE2 Poster Session II @InProceedings{Liu:2014:CECf, title = {A Multi-Swarm Particle Swarm Optimization with Orthogonal Learning for Locating and Tracking Multiple Optima in Dynamic Environments}, author = {Ruochen Liu and Xu Niu and Licheng Jiao}, pages = {754--761}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments, Particle swarm optimisation (PSO)}, abstract = { Due to the specificity and complexity of the dynamic optimisation problems (DOPs), those excellent static optimisation algorithms cannot be applied in these problems directly. So some special algorithms only for DOPs are needed. There is a multi-swarm algorithm with a better performance than others in DOPs, which uses a parent swarm to explore the search space and some child swarms to exploit promising areas found by the parent swarm. In addition, a static optimisation algorithm OLPSO is so attractive, which uses an orthogonal learning (OL) strategy to use previous search information (experience) more efficiently to predict the positions of particles and improve the convergence speed. In this paper, we bring the essence of OLPSO called OL strategy to the multi-swarm algorithm to improve its performance further. The experimental results conducted on different dynamic environments modelled by moving peaks benchmark show that the efficiency of this algorithm for locating and tracking multiple optima in dynamic environments is outstanding in comparison with other particle swarm optimisation models, including MPSO, a similar particle swarm algorithm for dynamic environments. }} @InProceedings{Liu:2014:CECg, title = {Regression Ensemble with {PSO} Algorithms Based Fuzzy Integral}, author = {James Liu and Yulin He and Yanxing Hu}, pages = {762--768}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary fuzzy systems}, abstract = { Similar to the ensemble learning for classification, regression ensemble also tries to improve the prediction accuracy through combining several ``weak'' estimators which are usually high-variance and thus unstable. In this paper, we propose a new scheme of fusing the weak Priestley-Chao Kernel Estimators (PCKEs) based on Choquet fuzzy integral, which differs from all the existing models of regressor fusion. The new scheme uses Choquet fuzzy integral to fuse several target outputs from different PCKEs, in which the optimal bandwidths are obtained with cross-validation criteria. The key of applying fuzzy integral to PCKE fusion is the determination of fuzzy measure. Considering the advantage of particle swarm optimisation (PSO) algorithm on convergence rate, we use three different PSO algorithms, i.e., standard PSO (SPSO), Gaussian PSO (GPSO) and GPSO with Gaussian jump (GPSOGJ), to determine the general and \${$\backslash$}lambda\$ fuzzy measures. The finally experimental results on 6 standard testing functions show that the new paradigm for regression ensemble based on fuzzy integral is more accurate and stable in comparison with any individual PCKE. This demonstrates the feasibility and effectiveness of our proposed regression ensemble model. }} @InProceedings{Jiang:2014:CEC, title = {An Improved Quantum-Behaved Particle Swarm Optimization Based on Linear Interpolation}, author = {Shouyong Jiang and Shengxiang Yang}, pages = {769--775}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Numerical optimisation}, abstract = { Quantum-behaved particle swarm optimisation (QPSO) has shown to be an effective algorithm for solving global optimisation problems that are of high complexity. This paper presents a new QPSO algorithm, denoted LI-QPSO, which employs a model-based linear interpolation method to strengthen the local search ability and improve the precision and convergence performance of the QPSO algorithm. In LI-QPSO, linear interpolation is used to approximate the objective function around a pre-chosen point with high quality in the search space. Then, local search is used to generate a promising trial point around this pre-chosen point, which is then used to update the worst personal best point in the swarm. Experimental results show that the proposed algorithm provides some significant improvements in performance on the tested problems. }} @InProceedings{Oh:2014:CEC, title = {Evolving Hierarchical Gene Regulatory Networks for Morphogenetic Pattern Formation of Swarm Robotics}, author = {Hyondong Oh and Yaochu Jin}, pages = {776--783}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Robotics, Robotics, Artificial ecology and artificial life}, abstract = { Morphogenesis, the biological developmental process of multicellular organisms, is a robust self-organising mechanism for pattern formation governed by gene regulatory networks (GRNs). Recent findings suggest that GRNs often show the use of frequently recurring patterns termed network motifs. Inspired by these biological studies, this paper proposes a morphogenetic approach to pattern formation for swarm robots to entrap targets based on an evolving hierarchical gene regulatory network (EH-GRN). The proposed EH-GRN consists of two layers: the upper layer is for adaptive pattern generation where the GRN model is evolved by basic network motifs, and the lower layer is responsible for driving robots to the target pattern generated by the upper layer. Obstacle information is introduced as one of environmental inputs along with that of targets in order to generate an adaptive pattern to unknown environmental changes. Besides, splitting or merging of multiple patterns resulting from target movement is addressed by the inherent feature of the upper layer and the \$k\$-means clustering algorithm. Numerical simulations have been performed for scenarios containing static/moving targets and obstacles to validate the effectiveness and benefit of the proposed approach for complex shape generation in dynamic environments. }} @InProceedings{Zheng:2014:CECb, title = {Avoiding Decoys in Multiple Targets Searching Problems Using Swarm Robotics}, author = {Zhongyang Zheng and Junzhi Li and Jie Li and Ying Tan}, pages = {784--791}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary games and multi-agent systems, Coevolution and collective behaviour}, abstract = { In this paper, we consider the target searching problems with a new type of the object: decoys which can be sensed exactly as targets but cannot be collected by the robots. In real-life applications, decoys are very common especially for swarm robots whose hardware should be designed as simple and cheap as possible. This inevitably brings errors and mistakes in the sensing results and the swarm may mistakenly sense certain kinds of environment objects as the target they are looking for. We proposed a simple cooperative strategy to solve this problem, comparing with a non-cooperative strategy as the baseline. The strategies work with other searching algorithms and provide schemes for avoiding decoys. Simulation results demonstrate that the cooperative strategy shares almost the same computation overload yet has better performance in iterations and especially visited times of decoys. The strategy shows great adaptivity to large scale problems and performs better when more decoys or robots exist in the simulation. }} @InProceedings{Liu:2014:CECh, title = {Particle Swarm Optimization for Integrity Monitoring in {BDS/DR} Based Railway Train Positioning}, author = {Jiang Liu and Bai-gen Cai and Jian Wang}, pages = {792--797}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Intelligent systems applications}, abstract = { Satellite navigation system, especially the BeiDou Navigation Satellite System (BDS), has become a significant resource for many transport branches. It is strongly required that BDS is applied in modern railway transportation systems to support the rapid development of Chinese railway infrastructure and services. Currently, the BDS is still in the developing period, and the existing resources are not sufficient to support integrity assurance for many safety-related railway applications. The aim of this paper is therefore to develop a novel integrity monitoring method for the BDS-based train positioning with assistance from the additional dead reckoning system. In this method, the raw measurements of sensors are fused with the Bayesian filtering, and the self-weight adaptive particle swarm optimisation with a combined objective function is involved to achieve an effective solution for the horizontal protection level which indicates the integrity capability. Field data are taken to validate effectiveness of the proposed solution and the advantages of the integrated particle fitness strategy. The implementation of this method will be positive for realising fault detection and isolation for a series of safety-related railway applications based on BDS. }} @InProceedings{Li:2014:CECd, title = {Learning and Evolution of Genetic Network Programming with Knowledge Transfer}, author = {Xianneng Li and Wen He and Kotaro Hirasawa}, pages = {798--805}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, genetic network programming, Representation and operators, Adaptive dynamic programming and reinforcement learning}, abstract = { Traditional evolutionary algorithms (EAs) generally starts evolution from scratch, in other words, randomly. However, this is computationally consuming, and can easily cause the instability of evolution. In order to solve the above problems, this paper describes a new method to improve the evolution efficiency of a recently proposed graph-based EA genetic network programming (GNP) by introducing {$\backslash$}emph\{knowledge transfer\} ability. The basic concept of the proposed method, named GNP-KT, arises from two steps: First, it formulates the knowledge by discovering abstract decision-making rules from source domains in a learning classifier system (LCS) aspect; Second, the knowledge is adaptively reused as advice when applying GNP to a target domain. A reinforcement learning (RL)-based method is proposed to automatically transfer knowledge from source domain to target domain, which eventually allows GNP-KT to result in better initial performance and final fitness values. The experimental results in a real mobile robot control problem confirm the superiority of GNP-KT over traditional methods. }} @InProceedings{Yang:2014:CEC, title = {An Improved {JADE} Algorithm for Global Optimization}, author = {Ming Yang and Zhihua Cai and Changhe Li and Jing Guan}, pages = {806--812}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution}, abstract = { In differential evolution (DE), the optimal value of the control parameters are problem-dependent. Many improved DE algorithms have been proposed with the aim of improving the effectiveness for solving general problems. As a very known adaptive DE algorithm, JADE adjusts the crossover probability CR of each individual by a norm distribution, in which the value of standard deviation is fixed, based on its historical record of success. The fixed and small standard deviation results in that the generated CR may not suitable for solving a problem. This paper proposed an improvement for the adaptation of CR, in which the standard deviation is adaptive. The diversity of values of CR was improved. This improvement was incorporated into the JADE algorithm and tested on a set of 25 scalable benchmark functions. The results showed that the adaptation of CR improved the performance of the JADE algorithm, particularly in comparisons with several other peer algorithms on high-dimensional functions. }} @InProceedings{Feng:2014:CEC, title = {Characterizing the Impact of Selection on the Evolution of Cooperation in Complex Networks}, author = {Shasha Feng and Shaolin Tan and Jinhu Lu}, pages = {813--818}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary games and multi-agent systems, Evolution strategies, Games}, abstract = { Cooperative behaviours are widespread in biological and social populations. Yet the evolution of cooperation is still a puzzle in evolutionary theory. Recent researches have indicated that complex interactions among individuals may promote the evolution of cooperation under weak selection. However, the selection effect on cooperation has not been completely understood. This paper aims to characterise the impact of selection on the emergence of cooperation in evolutionary dynamics on complex networks. By theoretical analysis and numerical simulation, it is found that selection favours defection over cooperation for the birth-death process, while it may favour cooperation over defection for the death-birth process. Furthermore, we come to the condition on which cooperation is dominant over defection. In particular, there exists an optimal selection intensity which favours cooperation the best for the death-birth process. The obtained results indicate that appropriate selection can promote the evolution of cooperation in structured populations under some circumstances. }} @InProceedings{Yu:2014:CECc, title = {A Tabu Search Heuristic for the Single Row Layout Problem with Shared Clearances}, author = {Meng Yu and Xingquan Zuo and Chase C. Murray}, pages = {819--825}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial optimisation, Heuristics, metaheuristics and hyper-heuristics}, abstract = { The single row layout problem is a common and well-studied practical facility layout problem. The problem seeks the arrangement of a fixed number of facilities along one row that minimises the objective of total material handling cost. In this paper, a single row layout problem with shared clearance between facilities is proposed. The shared additional clearance may be considered on one or both sides of each facility. To solve this problem tabu search is combined with a heuristic rule to solve problems of realistic size. Tabu search is used to find the sequence of facilities while the heuristic rule is determines the additional clearance for each facility. The proposed solution approach is applied to several problem instances involving 10, 20 and 30 facilities, and is compared against a popular mathematical programming solver (CPLEX). Computational results show that our approach is able to obtain high quality solutions and outperforms CPLEX under limited computational time for problems of realistic sizes. }} @InProceedings{Gao:2014:CEC, title = {A Weighting-Based Local Search Heuristic Algorithm for the Set Covering Problem}, author = {Chao Gao and Thomas Weise and Jinlong Li}, pages = {826--831}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Discrete and combinatorial optimisation}, abstract = { The Set Covering Problem (SCP) is NP-hard and has many applications. In this paper, we introduce a heuristic algorithm for SCPs based on weighting. In our algorithm, a local search framework is proposed to perturb the candidate solution under the best objective value found during the search, a weighting scheme and several search strategies are adopted to help escape from local optima and make the search more divergent. The effectiveness of our algorithm is evaluated on a set of instances from the OR-Library and Steiner triple systems. The experimental results show that it is very competitive, for it is able to find all the optima or best known results with very small run times on non-unicost instances from the OR-Library and outperforms two excellent solvers we have found in literature on the unicost instances from Steiner triple systems. Furthermore, it is conceptually simple and only needs one parameter to indicate the stopping criterion. }} @InProceedings{Schlueter:2014:CEC, title = {Parallelization for Space Trajectory Optimization}, author = {Martin Schlueter and Masaharu Munetomo}, pages = {832--839}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation, Parallel and Distributed Evolutionary Computation in the Cloud Era, Parallel and distributed algorithms}, abstract = { The impact of parallelisation on the optimisation process of space mission trajectories is investigated in this contribution. As space mission trajectory reference model, the well known Cassini1 benchmark, published by the European Space Agency (ESA), is considered and solved here with the MIDACO optimisation software. It can be shown that significant speed ups can be gained by applying parallelisation. }} @InProceedings{Jiang:2014:CECa, title = {Optimal Approximation of Stable Linear Systems with a Novel and Efficient Optimization Algorithm}, author = {Qiaoyong Jiang and Lei Wang and Xinhong Hei and Rong Fei and Dongdong Yang and Feng Zou and Hongye Li and Zijian Cao}, pages = {840--844}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics}, abstract = { Optimal approximation of linear system models is an important task in the controller design and simulation for complex dynamic systems. In this paper, we put forward a novel nature-based meta-heuristic method, called artificial raindrop algorithm, which is inspired from the phenomenon of natural rainfall, and apply it for optimal approximation for a stable linear system. It mimics the changing process of a raindrop, including the generation of raindrop, the descent of raindrop, the collision of raindrop, the flowing of raindrop and the updating of raindrop. Five corresponding operators are designed in the algorithm. Numerical experiment is carried on optimal approximation of a typical stable linear system in two fixed search intervals. The result demonstrates better performance of the proposed algorithm comparing with that of other five state-of-the-art optimisation algorithms. }} @InProceedings{Bolufe-Rohler:2014:CEC, title = {Extending Minimum Population Search Towards Large Scale Global Optimization}, author = {Antonio Bolufe-Rohler and Stephen Chen}, pages = {845--852}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Large Scale Global Optimisation, Large-scale problems}, abstract = { Minimum Population Search is a new metaheuristic specifically designed for multi-modal optimisation. Its core idea is to guarantee exploration in all dimensions of the search space with the smallest possible population. A small population increases the chances of convergence and the efficient use of function evaluations, an important consideration when scaling a search technique up towards large scale global optimisation. As the cost to converge to any local optimum increases in high dimensional search spaces, metaheuristics must focus more and more on gradient exploitation. To successfully maintain its balance between exploration and exploitation, Minimum Population Search uses thresheld convergence. Thresheld convergence can ensure that a search technique will perform a broad, unbiased exploration at the beginning and also have enough function evaluations allocated for proper convergence at the end. Experimental results show that Minimum Population Search outperforms Differential Evolution and Particle Swarm Optimisation on complex multi-modal fitness functions across a broad range of problem sizes. }} @InProceedings{Zhang:2014:CECc, title = {A New Penalty Function Method for Constrained Optimization Using Harmony Search Algorithm}, author = {Biao Zhang and Jun-hua Duan and Hong-yan Sang and Jun-qing Li and Hui Yan}, pages = {853--859}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { This paper proposes a novel penalty function measure for constrained optimisation using a new harmony search algorithm. In the proposed algorithm, a two-stage penalty is applied to the infeasible solutions. In the first stage, the algorithm can search for feasible solutions with better objective values efficiently. In the second stage, the algorithm can take full advantage of the information contained in infeasible individuals and avoid trapping in local optimum. In addition, for adapting to this method, a new harmony search algorithm is presented, which can keep a balance between exploration and exploitation in the evolution process. Numerical results of 13 benchmark problems show that the proposed algorithm performs more effectively than the ordinary methods for constrained optimisation problems. }} @InProceedings{Davendra:2014:CEC, title = {Scatter Search Algorithm with Chaos Based Stochasticity}, author = {Donald Davendra and Roman Senkerik and Ivan Zelinka and Michal Pluhacek}, pages = {860--866}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Emergent technologies, Evolutionary Computing with Deterministic Chaos}, abstract = { In this paper, we introduce a Scatter Search algorithm which is driven using a set of four chaos maps. The chaos maps of Tinkerbell, Delayed Logistics, Lozi and Burgers are used as chaotic pseudorandom number generators in the Scatter Search algorithm. These variants of the algorithm are used to solve the flowshop with blocking problem. The results are compared with the Mersenne Twister version of Scatter Search. The new chaos driven Scatter Search algorithm is shown to have superior performance when compared with state of the art heuristics in literature. }} @InProceedings{Akhmedova:2014:CEC, title = {Co-Operation of Biology Related Algorithms Meta-Heuristic in {ANN}-Based Classifiers Design}, author = {Shakhnaz Akhmedova and Eugene Semenkin}, pages = {867--872}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Coevolution and collective behaviour, Evolved neural networks}, abstract = { Meta-heuristic called Co-Operation of Biology Related Algorithms (COBRA), that has earlier demonstrated its usefulness on CEC'2013 real-valued optimisation competition benchmark, is applied to ANN-based classifiers design. The basic idea consists in representation of ANN's structure as a binary string and the use of the binary modification of COBRA for the ANN's structure selection. Neural network's weight coefficients represented as a string of real-valued variables are adjusted with the original version of COBRA. Four benchmark classification problems (two bank scoring problems and two medical diagnostic problems) are solved with this approach. Multilayered feed-forward ANNs with maximum 5 hidden layers and maximum 5 neurons on each layer are used. It means that ANN's structure optimal selection requires solving an optimisation problem with 100 binary variables. Fitness function calculation for each bit string requires solving an optimisation problem with up to 225 real-valued variables. Experiments showed that both variants of COBRA demonstrate high performance and reliability in spite of the complexity of solved optimisation problems. ANN-based classifiers developed in this way outperform many alternative methods on mentioned benchmark classification problems. The workability and usefulness of proposed meta-heuristic optimisation algorithms are confirmed. }} @InProceedings{Felipe:2014:CEC, title = {Scientific Algorithms for the Car Renter Salesman Problem}, author = {Denis Felipe and Elizabeth Ferreira Gouvea Goldbarg and Marco Cesar Goldbarg}, pages = {873--879}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Evolutionary programming}, abstract = { This paper presents the Scientific Algorithms, a new metaheuristics inspired in the scientific research process. The new method introduces the idea of theme to search the solution space of hard problems. The inspiration for this class of algorithms comes from the act of researching that comprises thinking, knowledge sharing and disclosing new ideas. The ideas of the new method are illustrated in the Travelling Salesman Problem. A computational experiment applies the proposed approach to a new variant of the Traveling Salesman Problem named Car Renter Salesman Problem. The results are compared to state-of-the-art algorithms for the latter problem. }} @InProceedings{Watanabe:2014:CEC, title = {A Proposal on Analysis Support System Based on Association Rule Analysis for Non-Dominated Solutions}, author = {Shinya Watanabe and Yuta Chiba and Masahiro Kanazaki}, pages = {880--887}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Multiobjective optimisation, Data mining}, abstract = { This paper presents a new analysis support system for analysing non-dominated solutions (NDSs) derived by evolutionary multi-criterion optimisation (EMO). The main features of the proposed system are to use association rule analysis and to perform a multi-granularity analysis based on a hierarchical tree of NDSs. The proposed system applies association rule analysis to the whole NDSs and derives association rules related to NDSs. And a hierarchical tree is created through our original association rule grouping that guarantees to keep at least one common features. Each node of a hierarchical tree corresponds to one group consisting of association rules and is fixed in position according to inclusion relations between groups. Since each group has some kinds of common features, the designer can analyse each node with previous knowledge of these common features. To investigate the characteristics and effectiveness of the proposed system, the proposed system is applied to the concept design problem of hybrid rocket engine (HRE) which has two objectives and six variable parameters. HRE separately stores two different types of thrust propellant unlike in the case of usual other rockets and the concept design problem of HRE has been provided by JAXA. The results of this application provided possible to analyse the trends and specifics contained in NDSs in an organised way unlike analysis approaches targeted at the whole NDSs. }} @InProceedings{Zhou:2014:CEC, title = {{GEAS}: A {GA-ES}-Mixed Algorithm for Parameterized Optimization Problems - Using {CLS} Problem as an Example}, author = {Xing Zhou and Wei Peng and Bo Yang}, pages = {888--894}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Discrete and combinatorial optimisation, Engineering applications}, abstract = { Parametrised optimisation problems (POPs) belong to a class of NP problems which are hard to be tackled by traditional methods. However, the relationship of the parameters (usually represented as \$k\$) makes a POP different from ordinary NP-complete problem in designing algorithms. In this paper, GEAS, an evolutionary computing algorithm (also can be seen as a framework) to solve POPs is proposed. This algorithm organically unifies genetic algorithm (GA) framework and the idea of evolutionary strategy (ES). It can maintain diversity while with a small population and has an intrinsic parallelism property:each individual in the population can solve a same problem that only has a different parameter. GEAS is delicately tested on an NP-complete problem, the {$\backslash$}textit \{Critical Link Set Problem\}. Experiment results show that GEAS can converge much faster and obtain more precise solution than GA which uses the same genetic operators. }} @InProceedings{Alvares:2014:CEC, title = {Application of Computational Intelligence for Source Code Classification}, author = {Marcos Alvares and Fernando Buarque and Tshilidzi Marwala}, pages = {895--902}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Classification, clustering and data analysis, Genetic algorithms, Real-world applications}, abstract = { Multi-language Source Code Management systems have been largely used to collaboratively manage software development projects. These systems represent a fundamental step in order to fully use communication enhancements by producing concrete value on the way people collaborate to produce more reliable computational systems. These systems evaluate results of analyses in order to organise and optimise source code. These analyses are strongly dependent on technologies (i.e. framework, programming language, libraries) each of them with their own characteristics and syntactic structure. To overcome such limitation, source code classification is an essential pre-processing step to identify which analyses should be evaluated. This paper introduces a new approach for generating content-based classifiers by using Evolutionary Algorithms. Experiments were performed on real world source code collected from more than 200 different open source projects. Results show us that our approach can be successfully used for creating more accurate source code classifiers. The resulting classifier is also expansible and flexible to new classification scenarios (opening perspectives for new technologies). }} @InProceedings{Hu:2014:CECa, title = {Genetic Algorithm with Spatial Receding Horizon Control for the Optimization of Facility Locations}, author = {Xiao-Bing Hu and Mark S Leeson}, pages = {903--909}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Combinatorial Optimisation, Large-scale problems}, abstract = { Inspired by the temporal receding horizon control in control engineering, this paper reports a novel spatial receding horizon control (SRHC) strategy to partition the facility location optimisation problem (FLOP), in order to reduce the complexity caused by the problem scale. Traditional problem partitioning methods can be viewed as a special case of the proposed SRHC, i.e., one-step-wide SRHC, whilst the method in this paper is a generalised N-step-wide SRHC, which can make a better use of global information of the route network where a given number of facilities need to be set up. With SRHC to partition the FLOP, genetic algorithm (GA) is integrated as optimiser to resolve the partitioned problem within each spatial receding horizon. On one hand, SRHC helps to improve the scalability of GA. On the other, the population feature of GA helps to reduce the shortsighted performance of SRHC. The effectiveness and efficiency of the reported SRHC and GA for the FLOP are demonstrated by comparative simulation results. }} @InProceedings{Reps:2014:CEC, title = {Tuning a Multiple Classifier System for Side Effect Discovery Using Genetic Algorithms}, author = {Jenna Reps and Uwe Aickelin and Jonathan Garibaldi}, pages = {910--917}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence in Bioinformatics}, abstract = { In previous work, a novel supervised framework implementing a binary classifier was presented that obtained excellent results for side effect discovery. Interestingly, unique side effects were identified when different binary classifiers were used within the framework, prompting the investigation of applying a multiple classifier system. In this paper we investigate tuning a side effect multiple classifying system using genetic algorithms. The results of this research show that the novel framework implementing a multiple classifying system trained using genetic algorithms can obtain a higher partial area under the receiver operating characteristic curve than implementing a single classifier. Furthermore, the framework is able to detect side effects efficiently and obtains a low false positive rate. }} @InProceedings{Zhang:2014:CECd, title = {Cooperation with Potential Leaders in Evolutionary Game Study of Networking Agents}, author = {Jianlei Zhang and Chunyan Zhang and Tianguang Chu and Ming Cao}, pages = {918--923}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Games, Coevolution and collective behaviour, Evolution strategies}, abstract = { Increasingly influential leadership is significant to the cooperation and success of human societies. However, whether and how leaders emerge among evolutionary game players still remain less understood. Here, we study the evolution of potential leaders in the framework of evolutionary game theory, adopting the prisoner's dilemma and snowdrift game as metaphors of cooperation between unrelated individuals. We find that potential leaders can spontaneously emerge from homogeneous populations along with the evolution of cooperation, demonstrated by the result that a minority of agents spread their strategies more successfully than others and guide the population behaviour, irrespective of the applied games. In addition, the phenomenon just described can be observed more notably in populations situated on scale free networks, and thus implies the relevance of heterogeneous networks for the possible emergence of leadership in the proposed system. Our results underscore the importance of the study of leadership in the population indulging in evolutionary games. }} @InProceedings{Duan:2014:CEC, title = {Multi-Objective Optimization Model Based on Steady Degree for Teaching Building Evacuation}, author = {Pengfei Duan and Shengwu Xiong and Zhongbo Hu and Qiong Chen and Xinlu Zhong}, pages = {924--929}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Multi-objective evolutionary algorithms, Discrete and combinatorial optimisation}, abstract = { In this paper, the process of evacuation in teaching building is considered. The concept of steady degree based on cellular automata and potential field is introduced and it can describe the behaviour tendency of evacuees during the evacuation process. With the help of steady degree, the model simulates the indoor evacuation behaviour. To reduce the congestion and evacuation time, a multi-objective optimisation model considering steady degree and evacuation clearance time is proposed. Finally, an experiment in the Teaching Building No.1 of Wuhan University of Technology is carried out. The results show that this model can reduce the clearance time of emergency evacuation in teaching building compared to other models. }} @InProceedings{Bello-Orgaz:2014:CEC, title = {Evolutionary Clustering Algorithm for Community Detection Using Graph-Based Information}, author = {Gema Bello-Orgaz and David Camacho}, pages = {930--937}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis}, abstract = { The problem of community detection has become highly relevant due to the growing interest in social networks. The information contained in a social network is often represented as a graph. The idea of graph partitioning of graph theory can be apply to split a graph into node groups based on its topology information. In this paper the problem of detecting communities within a social network is handled applying graph clustering algorithms based on this idea. The new approach proposed is based on a genetic algorithm. A new fitness function has been designed to guide the clustering process combining different measures of network topology (Density, Centralisation, Heterogeneity, Neighbourhood, Clustering Coefficient). These different network measures have been experimentally tested using a real-world social network. Experimental results show that the proposed approach is able to detect communities and the results obtained in previous work have been improved. }} @InProceedings{Nishiyama:2014:CEC, title = {Applying Conversion Matrix to Robots for Imitating Motion Using Genetic Algorithms}, author = {Mari Nishiyama and Hitoshi Iba}, pages = {938--944}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Evolutionary Robotics, Real-world applications}, abstract = { In this paper, we propose a method using a genetic algorithm (GA) for motion imitation between two different types of humanoid robots. Although motion imitation between humans and robots has been a popular research topic for a long time, the imitation between different types of robots still remains an unsolved task. The selection of the correct joint angles is critical for robot motion. However, different robots have different anatomies, with each joint's position and movable range uniquely defined for each type of robot. This discrepancy is an obstacle when converting a motion to another type of robot. The proposed method uses a genetic algorithm in order to find the conversion matrix needed to map one robot's joint angles to joint angles of another robot. This is done with two objectives in mind; one is to reduce the difference between the sample imitation and the converted imitation. The other one is to keep the stability. Two experiments were conducted; one stable and one unstable experiment. The experiments were made with two different types of robots in a simulation environment. The stable experiment showed a concordance rate of 93.7\% with the test motion. The imitation also tested with the real robot and succeeded to keep standing. In the unstable experiment, the student robot keeps its balance for most of the simulation time. It showed a concordance rate of 95.5\%, which is slightly higher than that in the stable experiment. These results show great promise for the proposed method as a way to realise motion imitation between different types of robots. }} @InProceedings{Manfrini:2014:CEC, title = {Optimization of Combinational Logic Circuits Through Decomposition of Truth Table and Evolution of Sub-Circuits}, author = {Francisco Manfrini and Helio Barbosa and Heder Bernadino}, pages = {945--950}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Evolvable hardware and software}, abstract = { In this work, a genetic algorithm was used to design combinational logic circuits (CLCs), with the goal of minimising the number of logic elements in the circuit. A new coding for circuits is proposed using a multiplexer (MUX) at the output of the circuit. This MUX divides the truth table into two distinct parts, with the evolution occurring in three sub-circuits connected to the control input and the two data inputs of the MUX. The methodology presented was tested with some benchmark circuits. The results were compared with those obtained using traditional design methods, as well as the results found in other articles, which used different heuristics to design CLCs. }} @InProceedings{Huynh-Thi-Thanh:2014:CEC, title = {Reordering Dimensions for Radial Visualization of Multidimensional Data - A Genetic Algorithms Approach}, author = {Binh Huynh Thi Thanh and Long Tran Van and Hoai Nguyen Xuan and Anh Nguyen Duc and Truong Pham Manh}, pages = {951--958}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Intelligent systems applications, Discrete and combinatorial optimisation}, abstract = { In this paper, we propose a Genetic Algorithm (GA) for solving the problem of dimensional ordering in Radial Visualisation (Radviz) systems. The Radviz is a non-linear projection of high-dimensional data set onto two dimensional space. The order of dimension anchors in the Radviz system is crucial for the visualisation quality. We conducted experiments on five common data sets and compare the quality of solutions found by GA and those found by the other well-known methods. The experimental results show that the solutions found by GA for these tested data sets are very competitive having better cost values than almost all solutions found by other methods. This suggests that GA could be a good approach to solve the problem of dimensional ordering in Radviz. }} @InProceedings{Silva:2014:CEC, title = {An Evolutionary Approach for Combining Results of Recommender Systems Techniques Based on Collaborative Filtering}, author = {Edjalma Queiroz Silva and Celso Goncalves Camilo-Junior and Luiz Mario L. Pascoal and Thierson Couto Rosa}, pages = {959--966}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Real-world applications, Intelligent systems applications}, abstract = { Recommendation systems work as a counsel or, behaving in such a way to guide people in the discovery of products of interest. There are various techniques and approaches in the literature that enable generating recommendations. This is interesting because it emphasises the diversity of options; on the other hand, it can cause doubt to the system designer about which is the best technique to use. Each of these approaches has particularities and depends on the context to be applied. Thus, the decision to choose among techniques become complex to be done manually. This article proposes an evolutionary approach for combining results of recommendation techniques in order to automate the choice of techniques and get fewer errors in recommendations. To evaluate the proposal, experiments were performed with a dataset from MovieLens and some of Collaborative Filtering techniques. The results show that the combining methodology proposed in this paper performs better than any one of collaborative filtering technique separately in the context addressed. The improvement varies from 9.02\% to 48.21\% depending on the technique and the experiment executed. }} % Special Session: TuE2-1 Nature-Inspired Constrained Optimisation @InProceedings{Bu:2014:CEC, title = {Differential Evolution with a Species-Based Repair Strategy for Constrained Optimization}, author = {Chenyang Bu and Wenjian Luo and Tao Zhu}, pages = {967--974}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Nature-Inspired Constrained Optimisation, Constraint handling, Differential evolution}, abstract = { Evolutionary Algorithms (EAs) with gradient-based repair, which uses the gradient information of the constraints set, have been proved to be effective. It is known that it would be time-consuming if all infeasible individuals are repaired. Therefore, so far the infeasible individuals to be repaired are randomly selected from the population and the strategy of choosing individuals to be repaired has not been studied yet. In this paper, the Species-based Repair Strategy (SRS) is proposed to select representative infeasible individuals instead of the random selection for gradient-based repair. The proposed SRS strategy has been applied to eDEag which repairs the random selected individuals using the gradient-based repair. The new algorithm is named SRS-eDEag. Experimental results show that SRS-eDEag outperforms eDEag in most benchmarks. Meanwhile, the number of repaired individuals is reduced markedly. }} @InProceedings{Ameca-Alducin:2014:CEC, title = {Differential Evolution with Combined Variants for Dynamic Constrained Optimization}, author = {Maria-Yaneli Ameca-Alducin and Efren Mezura-Montes and Nicandro Cruz-Ramirez}, pages = {975--982}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Nature-Inspired Constrained Optimisation}, abstract = { In this work a differential evolution algorithm is adapted to solve dynamic constrained optimisation problems. The approach is based on a mechanism to detect changes in the objective function and/or the constraints of the problem so as to let the algorithm to promote the diversity in the population while pursuing the new feasible optimum. This is made by combining two popular differential evolution variants and using a memory of best solutions found during the search. Moreover, random-immigrants are added to the population at each generation and a simple hill-climber-based local search operator is applied to promote a faster convergence to the new feasible global optimum. The approach is compared against other recently proposed algorithms in an also recently proposed benchmark. The results show that the proposed algorithm provides a very competitive performance when solving different types of dynamic constrained optimisation problems. }} @InProceedings{Singh:2014:CEC, title = {Solving Problems with a Mix of Hard and Soft Constraints Using Modified Infeasibility Driven Evolutionary Algorithm ({IDEA-M})}, author = {Hemant Singh and Md. Asafuddoula and Tapabrata Ray}, pages = {983--990}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling, Nature-Inspired Constrained Optimisation, Numerical optimisation}, abstract = { Most optimisation problems in the field of engineering design involve constraints. These constraints are often due to statutory requirements (e.g. safety, physical laws, user requirements/functionality) and/or limits imposed on time and resources. Population based stochastic optimisation algorithms are a preferred choice for solving design optimisation problems due to their ability to deal with nonlinear black-box functions. Having a good constraint handling technique embedded within the algorithm is imperative for its good performance. With the final aim of achieving feasible optimum solutions, feasibility first techniques, i.e., those which prefer feasible solutions over infeasible, have been commonly used in the past. However, in recent studies more emphasis has been laid on intelligent use of infeasible solutions (instead of their indiscreet rejection) during the course of optimisation; particularly because optimum solutions often lie on the constraint boundary. The preservation of good infeasible solutions in the population is likely to improve the convergence in constricted or disconnected feasible regions. In addition, it provides a set of marginally infeasible solutions for trade-off considerations. However, in the case of a problem consisting of a mix of hard (non-negotiable) and soft (negotiable) constraints, such trade-off solutions are practically useful if they violate the soft constraints only. In this paper, previously introduced Infeasibility Driven Evolutionary Algorithm (IDEA) is modified to deliver solutions which strictly satisfy the hard constraints and offer trade off solutions with respect to the soft constraints. The performance of the algorithm is demonstrated on three benchmark problems. }} @InProceedings{Hamza:2014:CEC, title = {Differential Evolution with a Constraint Consensus Mutation for Solving Optimization Problems}, author = {Noha Hamza and Ruhul Sarker and Daryl Essam}, pages = {991--997}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Nature-Inspired Constrained Optimisation}, abstract = { in the literature, a considerable number of mutation operators have been proposed, which are the key search operators in differential evolution algorithm for solving optimisation problems. Although those operators were developed in the context of unconstrained optimisation, they were widely used in constrained optimisation. However, those operators did not contain any mechanism that would reduce the constraint violation in the search process. Therefore, in this paper, a new mutation operator based on the constraint consensus method is proposed, which can help infeasible points reach the feasible region quickly. The algorithm is tested on the CEC2010 constrained benchmark problems. The experimental results show that the proposed algorithm is able to obtain better solutions in comparison with the state-of-the-art algorithms. }} @InProceedings{Poole:2014:CEC, title = {Constraint Handling in Agent-Based Optimization by Independent Sub-Swarms}, author = {Daniel Poole and Christian Allen and Thomas Rendall}, pages = {998--1005}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling}, abstract = { Agent-based optimisation algorithms are an effective means of solving global optimisation problems with design spaces containing multiple local minima, however, modifications have to be made to such algorithms to be able to solve constrained optimisation problems. The gravitational search algorithm (GSA) is an efficient and effective agent-based method, however, the idea of global transfer of data that is key to the algorithm's success prohibits coupling of many state-of-the-art methods for handling constraints. Hence, a robust method, called separation-sub-swarm (3S) has been developed specifically for use with GSA by exploiting but also accommodating the global transfer of data that occurs in GSA, however it can also act as an entirely black-box module so is generally applicable. This newly developed 3S method has been shown to be efficient and effective at optimising a suite of constrained analytical test functions using GSA. }} @InProceedings{Elsayed:2014:CEC, title = {United Multi-Operator Evolutionary Algorithms}, author = {Saber Elsayed and Ruhul Sarker and Daryl Essam}, pages = {1006--1013}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Nature-Inspired Constrained Optimisation, Differential Evolution: Past, Present and Future, Numerical optimisation}, abstract = { Multi-method and multi-operator evolutionary algorithms (EAs) have shown superiority to any single EAs with a single operator. To further improve the performance of such algorithms, in this research study, a united multi-operator EAs framework is proposed, in which two EAs, each with multiple search operators, are used. During the evolution process, the algorithm emphasises on the best performing multi-operator EA, as well as the search operator. The proposed algorithm is tested on a well-known set of constrained problems with 10D and 30D. The results show that the proposed algorithm scales well and is superior to the state-of-the-art algorithms, especially for the 30D test problems }} % Special Session: TuE2-2 Computational Intelligence in Bioinformatics @InProceedings{Nobile:2014:CEC, title = {A Memetic Hybrid Method for the Molecular Distance Geometry Problem with Incomplete Information}, author = {Marco S. Nobile and Andrea G. Citrolo and Paolo Cazzaniga and Daniela Besozzi and Giancarlo Mauri}, pages = {1014--1021}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {GA, PSO, Computational Intelligence in Bioinformatics}, abstract = { The definition of computational methodologies for the inference of molecular structural information plays a relevant role in disciplines as drug discovery and metabolic engineering, since the functionality of a biochemical molecule is determined by its three-dimensional structure. In this work, we present an automatic methodology to solve the Molecular Distance Geometry Problem, that is, to determine the best three-dimensional shape that satisfies a given set of target inter-atomic distances. In particular, our method is designed to cope with incomplete distance information derived from Nuclear Magnetic Resonance measurements. To tackle this problem, that is known to be NP-hard, we present a memetic method that combines two soft-computing algorithms: Particle Swarm Optimisation and Genetic Algorithms, with a local search approach, to improve the effectiveness of the crossover mechanism. We show the validity of our method on a set of reference molecules with a length ranging from 402 to 1003 atoms. }} @InProceedings{Thompson:2014:CEC, title = {{GAMI-CRM}: Using De Novo Motif Inference to Detect Cis-Regulatory Modules}, author = {Jeffrey A. Thompson and Clare Bates Congdon}, pages = {1022--1029}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence in Bioinformatics, Biometrics, bioinformatics and biomedical applications}, abstract = { In this work, we extend GAMI (Genetic Algorithms for Motif Inference), a de novo motif inference system, to find sets of motifs that may function as part of a cis-regulatory module (CRM) using a comparative genomics approach. Evidence suggests that most transcription factors binding sites are part of a CRM, so our new approach is expected to yield stronger candidates for de novo inference of candidate regulatory elements and their combinatorial regulation of genes. Thanks to our genetic algorithms based approach, we are able to search relatively large input sequences (100,000nt or longer). Most current computational approaches to identifying candidate CRMs depend on foreknowledge of the processes that the genes they regulate are involved in. In comparison with one leading method, Cluster-Buster, our prototype de novo approach, which we call GAMI-CRM, performed well, suggesting that GAMI-CRM will be particularly useful in predicting CRMs for genes whose interactions are poorly understood. }} @InProceedings{Pang:2014:CEC, title = {An Immune Network Approach to Learning Qualitative Models of Biological Pathways}, author = {Wei Pang and George Coghill}, pages = {1030--1037}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Artificial immune systems, Biometrics, bioinformatics and biomedical applications, Real-world applications}, abstract = { In this paper we continue the research on learning qualitative differential equation (QDE) models of biological pathways building on previous work. In particular, we adapt opt-AiNet, an immune-inspired network approach, to effectively search the qualitative model space. To improve the performance of opt AiNet on the discrete search space, the hypermutation operator has been modified, and the affinity between two antibodies has been redefined. In addition, to accelerate the model verification process, we developed a more efficient Waltz-like inverse model checking algorithm. Finally, a Bayesian scoring function is incorporated into the fitness evaluation to better guide the search. Experimental results on learning the detoxification pathway of Methylglyoxal with various hypothesised hidden species validate the proposed approach, and indicate that our opt-AiNet based approach outperforms the previous CLONALG based approach on qualitative pathway identification. }} @InProceedings{Chen:2014:CECa, title = {Multi-Dimensional Scaling and {MODELLER}-Based Evolutionary Algorithms for Protein Model Refinement}, author = {Yan Chen and Yi Shang and Dong Xu}, pages = {1038--1045}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Biometrics, bioinformatics and biomedical applications, Computational Intelligence in Bioinformatics}, abstract = { Protein structure prediction, i.e., computationally predicting the three dimensional structure of a protein from its primary sequence, is one of the most important and challenging problems in bioinformatics. Model refinement is a key step in the prediction process, where improved structures are constructed based on a pool of initially generated models. Since the refinement category was added to the biennial Critical Assessment of Structure Prediction (CASP) in 2008, CASP results show that it is a challenge for existing model refinement methods to improve model quality consistently. This paper presents three evolutionary algorithms for protein model refinement, in which multidimensional scaling (MDS), the MODELLER software, and a hybrid of both are used as crossover operators, respectively. The MDS-based method takes a purely geometrical approach and generates a child model by combining the contact maps of multiple parents. The MODELLER-based method takes a statistical and energy minimisation approach, and uses the remodelling module in MODELLER program to generate new models from multiple parents. The hybrid method first generates models using the MDS-based method and then run them through the MODELLER-based method, aiming at combining the strength of both. Promising results have been obtained in experiments using CASP datasets. The MDS-based method improved the best of a pool of predicted models in terms of the global distance test score (GDT-TS) in 9 out of 16 test targets. }} @InProceedings{Chowdhury:2014:CEC, title = {A Modified Bat Algorithm to Predict Protein-Protein Interaction Network}, author = {Archana Chowdhury and Pratyusha Rakshit and Amit Konar and Atulya Nagar}, pages = {1046--1053}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Biometrics, bioinformatics and biomedical applications, Evolutionary programming, Numerical optimisation}, abstract = { This paper provides a novel approach to predict the Protein-Protein Interaction (PPI) network using a modified version of the Bat Algorithm. The attractive trait of the proposed approach is that it attempts to analyse the impact of physicochemical properties, structural features and evolutionary relationship of proteins, to predict the PPI network. Computer simulations reveal that our proposed method effectively predicts the PPI of Saccharomyces Cerevisiae with a sensitivity of (0.85) and specificity of (0.87) and outperforms other state-of-art methodologies. }} @InProceedings{Peterson:2014:CEC, title = {Evolutionary Algorithms Applied to Likelihood Function Maximization During {Poisson}, Logistic, and {Cox} Proportional Hazards Regression Analysis}, author = {Leif Peterson}, pages = {1054--1061}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence in Bioinformatics, Particle swarm optimization (PSO), Heuristics, metaheuristics and hyper-heuristics}, abstract = { Metaheuristics based on genetic algorithms (GA), covariance matrix self-adaptation evolution strategies (CMSA-ES), particle swarm optimization (PSO), and ant colony optimization (ACO) were used for minimizing deviance for Poisson regression and maximising the log-likelihood function for logistic regression and Cox proportional hazards regression. We observed that, in terms of regression coefficients, CMSA-ES and PSO metaheuristics were able to obtain solutions that were in better agreement with Newton-Raphson (NR) when compared with GA and ACO. The rate of convergence to the NR solution was also faster for CMSA-ES and PSO when compared with ACO and GA. Overall, CMSA-ES was the best-performing method used. Key factors which strongly influence performance are multicollinearity, shape of the log-likelihood gradient, and positive definiteness of the Hessian matrix. }} % Special Session: TuE2-3 Single Objective Numerical Optimisation I @InProceedings{Elsayed:2014:CECa, title = {A Surrogate-Assisted Differential Evolution Algorithm with Dynamic Parameters Selection for Solving Expensive Optimization Problems}, author = {Saber Elsayed and Tapabrata Ray and Ruhul Sarker}, pages = {1062--1068}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { In this paper, a surrogate-assisted differential evolution (DE) algorithm is proposed to solve the computationally expensive optimisation problems. In it, the Kriging model is used to approximate the objective function, while DE employs a mechanism to dynamically select the best performing combinations of parameters (amplification factor, crossover rate and population size). The performance of the algorithm is tested on the WCCI2014 competition on expensive single objective optimisation problems. The experimental results demonstrate that the proposed algorithm has the ability to obtain good solutions. }} @InProceedings{Singh:2014:CECa, title = {A Hybrid Surrogate Based Algorithm ({HSBA}) to Solve Computationally Expensive Optimization Problems}, author = {Hemant Singh and Amitay Isaacs and Tapabrata Ray}, pages = {1069--1075}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Numerical optimisation}, abstract = { Engineering optimisation problems often involve multiple objectives and constraints that are computed via computationally expensive numerical simulations. While the severe nonlinearity of the objective/constraint functions demand the use of population based searches (e.g. Evolutionary Algorithms), such algorithms are known to require numerous function evaluations prior to convergence and hence may not be viable in their native form. On the other hand, gradient based algorithms are fast and effective in identifying local optimum, but their performance is dependent on the starting point. In this paper, a hybrid algorithm is presented, which exploits the benefits offered by population based scheme, local search and also surrogate modelling to solve optimisation problems with limited computational budget. The performance of the algorithm is reported on the benchmark problems designed for CEC 2014 Special Session and Competition on Single Objective Real-Parameter Numerical Optimisation. }} @InProceedings{Biswas:2014:CEC, title = {Evaluating the Performance of Group Counseling Optimizer on {CEC 2014} Problems for Computational Expensive Optimization}, author = {Subhodip Biswas and Mohammad A. Eita and Swagatam Das and Athanasios V. Vasilakos}, pages = {1076--1083}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { Group Counselling Optimiser (GCO) is a recently proposed population-based metaheuristics that simulates the ability of human beings to solve problems through counselling within a group and is motivated by the fact that the human thinking ability is often predicted to be the most rational. This research article examines the performance of GCO on the benchmark test suite designed for the CEC 2014 Competition for Computational Expensive Optimisation. Experimental results on 24 black-box optimisation problems (8 test problems with 10, 20 and 30 dimensions) have been tabulated along with the algorithm complexity metrics. Additionally we investigate the parametric behaviour of GCO based on these test instances. }} @InProceedings{Erlich:2014:CEC, title = {Solving the {IEEE-CEC 2014} Expensive Optimization Test Problems by Using Single-Particle {MVMO}}, author = {Istvan Erlich and Jose L. Rueda and Sebastian Wildenhues}, pages = {1084--1091}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { Mean-Variance Mapping Optimisation (MVMO) constitutes an emerging heuristic optimisation algorithm, whose evolutionary mechanism adopts a single parent-offspring pair approach along with a normalised range of the search space for all optimisation variables. Besides, MVMO is characterised by an archive of n-best solutions from which the unique mapping function defined by the mean and variance of the optimisation variables is derived. The algorithm proceeds by projecting randomly selected variables onto the corresponding mapping function that guides the solution towards the best set achieved so far. Despite the orientation on the best solution the algorithm keeps on searching globally. This paper provides an evaluation of the performance of MVMO when applied for the solution of computationally expensive optimisation problems. Experimental tests, conducted on the IEEE-CEC 2014 optimisation test bed, highlight the capability of the MVMO to successfully tackle different complex problems within a reduced number of allowed function evaluations. }} @InProceedings{Krityakierne:2014:CEC, title = {{SO-MODS}: Optimization for High Dimensional Computationally Expensive Multi-Modal Functions with Surrogate Search}, author = {Tipaluck Krityakierne and Juliane Mueller and Christine Shoemaker}, pages = {1092--1099}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { SO-MODS is a new algorithm that combines surrogate global optimisation methods with local search. SO-MODS is an extension of prior algorithms that sought to find near optimal solutions for computationally very expensive functions for which the number of allowable evaluations is strictly limited. The global search method in SO-MODS perturbs the best point found so far in order to find a new sample point. The number of decision variables being perturbed is dynamically adjusted in each iteration in order to be more effective for higher dimensional problems. The procedure for dynamically changing the dimensions perturbed is drawn from earlier work on the DYCORS algorithm. We use a cubic radial basis function as surrogate model and investigate two approaches to improve the solution accuracy. The numerical results show that SO-MODS is able to reduce the objective function value dramatically with just a few hundred evaluations even for 30-dimensional problems. The local search is then able to reduce the objective function value further. }} % Special Session: TuE2-4 Data Mining and Machine Learning Meet Evolutionary Computation @InProceedings{Rosales-Perez:2014:CEC, title = {An Evolutionary Multi-Objective Approach for Prototype Generation}, author = {Alejandro Rosales-Perez and Hugo Jair Escalante and Carlos A. Coello Coello and Jesus A. Gonzalez and Carlos A. Reyes-Garcia}, pages = {1100--1107}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Data Mining and Machine Learning Meet Evolutionary Computation, Classification, clustering and data analysis}, abstract = { k-NN is one of the most popular and effective models for pattern classification. However, it has two main drawbacks that hinder the application of this method for large data sets: (1) the whole training set has to be stored in memory, and (2) for classifying a test pattern it has to be compared to all other training instances. In order to overcome these shortcomings, prototype generation (PG) methods aim to reduce the size of the training set while maintaining or increasing the classification performance of k-NN. Accordingly, most PG methods aim to generate instances that try to maximise classification performance. Nevertheless, in most cases, the reduction objective is only implicitly optimised. This paper introduces EMOPG, a novel approach to PG based on multi-objective optimisation that explicitly optimises both objectives: accuracy and reduction. Under EMOPG, prototypes are initialised with a subset of training instances selected through a tournament, according to a weighting term. A multi-objective evolutionary algorithm, PAES (Pareto Archived Evolution Strategy), is implemented to adjust the position of the initial prototypes. The optimisation process aims to simultaneously maximise the classification performance of prototypes while reducing the number of instances with respect to the training set. A strategy for selecting a single solution from the set of non-dominated solutions is proposed. We evaluate the performance of EMOPG using a suite of benchmark data sets and compare the performance of our proposal with respect to the one obtained by alternative techniques. Experimental results show that our proposed method offers a better trade-off between accuracy and reduction than other methods. }} @InProceedings{Cheng:2014:CEC, title = {Use {EMO} to Protect Sensitive Knowledge in Association Rule Mining by Removing Items}, author = {Peng Cheng and Jeng-Shyang Pan and Chun-Wei Lin}, pages = {1108--1115}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Data Mining and Machine Learning Meet Evolutionary Computation, Multi-objective evolutionary algorithms, Data mining}, abstract = { When people use data mining techniques to discover useful knowledge behind large database; they also have the requirement to preserve some information so as not to be mined out, such as sensitive frequent item sets, rules, classification tree and the like. A feasible way to address this problem is to sanitise the database to conceal sensitive information. In this paper, we focus on privacy preserving in association rule mining. In light of the trade off between hiding sensitive rules and disclosing non-sensitive ones during hiding process, we tackle this problem from a point view of multi-objective optimisation. A novel association rule hiding approach was proposed based on evolutionary multi-objective optimisation (EMO) algorithm. It adopted the model of hiding sensitive rules by deleting some transactions in database. Three side effects, including sensitive rules not hidden, non-sensitive lost rules and spurious rules were formulated as objectives to be minimised. EMO algorithm is used to find a suitable subset of transactions to remove so that the three side effects are minimised. Experiment results were reported to show the effectiveness of the proposed approach. }} @InProceedings{Debie:2014:CEC, title = {An Online Evolutionary Rule Learning Algorithm with Incremental Attribute Discretization}, author = {Essam Debie and Kamran Shafi and Kathryn Merrick and Chris Lokan}, pages = {1116--1123}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Data Mining and Machine Learning Meet Evolutionary Computation, Data mining, Classification, clustering and data analysis}, abstract = { Classification rule induction involves two main processes: finding the optimal conjuncts (attribute intervals or attribute-value pairs) and their combination (disjuncts or rules) to classify different concepts in the data. The evolutionary rule learning approaches employ an evolutionary algorithm, such as a genetic algorithm, to perform both these search operations simultaneously. This approach often leads to significant problems including population bloating and stalled evolutionary search in real-valued attribute problems, especially with higher dimensions. In this paper, we present an online evolutionary rule learning approach referred to as ERL-AID that decouples the above search processes and employs a discrimination algorithm that works on the attribute space and a genetic algorithm to combine the discredited attributes into appropriate classification rules. ERL-AID applies a sliding window approach to process inputs in an online fashion. The proposed system is able to produce compact rule sets with competitive performance and could scale to higher dimensions. The experimental results show the competitiveness of our algorithm. }} @InProceedings{Yexing:2014:CEC, title = {An External Archive Guided Multiobjective Evolutionary Approach Based on Decomposition for Continuous Optimization}, author = {Li Yexing and Cai Xinye and Fan Zhun and Zhang Qingfu}, pages = {1124--1130}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Data Mining and Machine Learning Meet Evolutionary Computation, Multi-objective evolutionary algorithms}, abstract = { In this paper, we propose a decomposition based multiobjective evolutionary algorithm that extracts information from an external archive to guide the evolutionary search for continuous optimisation problem. The proposed algorithm used a mechanism to identify the promising regions(subproblems) through learning information from the external archive to guide evolutionary search process. In order to demonstrate the performance of the algorithm, we conduct experiments to compare it with other decomposition based approaches. The results validate that our proposed algorithm is very competitive. }} @InProceedings{Bourennani:2014:CEC, title = {Multi-Objective Differential Evolution with Leadership Enhancement ({MODEL})}, author = {Farid Bourennani and Shahryar Rahnamayan and Greg F. Naterer}, pages = {1131--1138}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Data Mining and Machine Learning Meet Evolutionary Computation, Self-adaptation in evolutionary computation}, abstract = { Differential Evolution (DE) has been successfully used to solve various complex optimisation problems; however, it can suffer depending of the complexity of the problem from slow convergence due to its iterative process. The use of the leadership concept was efficiently used for the acceleration of Particle Swarm Optimisation (PSO) in a single-objective space. The generalisation of the leadership concept in multi-objective space is not trivial. Furthermore, despite the efficiency of using the leadership concept, a limited number of multi-objective metaheuristics use it. To address these challenges, this paper incorporates the concept of leadership in a multi-objective variant of DE by introducing it into the mutation scheme. The preliminary results are promising as MODEL outperformed the parent algorithm GDE3 and showed the highest accuracy when compared with seven other algorithms. }} @InProceedings{Bandaru:2014:CEC, title = {On the Performance of Classification Algorithms for Learning {Pareto}-Dominance Relations}, author = {Sunith Bandaru and Amos Ng and Kalyanmoy Deb}, pages = {1139--1146}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Meta-modelling and surrogate models, Data Mining and Machine Learning Meet Evolutionary Computation}, abstract = { Multi-objective evolutionary algorithms (MOEAs) are often criticised for their high-computational costs. This becomes especially relevant in simulation-based optimisation where the objectives lack a closed form and are expensive to evaluate. Over the years, meta-modelling or surrogate modelling techniques have been used to build inexpensive approximations of the objective functions which reduce the overall number of function evaluations (simulations). Some recent studies however, have pointed out that accurate models of the objective functions may not be required at all since evolutionary algorithms only rely on the relative ranking of candidate solutions. Extending this notion to MOEAs, algorithms which can 'learn' Pareto dominance relations can be used to compare candidate solutions under multiple objectives. With this goal in mind, in this paper, we study the performance of ten different off-the-shelf classification algorithms for learning Pareto-dominance relations in the ZDT test suite of benchmark problems. We consider prediction accuracy and training time as performance measures with respect to dimensionality and skewness of the training data. Being a preliminary study, this paper does not include results of integrating the classifiers into the search process of MOEAs. }} % Session: WeE1-1 Multi-Objective Evolutionary Algorithms I @InProceedings{Purshouse:2014:CEC, title = {A Review of Hybrid Evolutionary Multiple Criteria Decision Making Methods}, author = {Robin C. Purshouse and Kalyanmoy Deb and Maszatul M. Mansor and Sanaz Mostaghim and Rui Wang}, pages = {1147--1154}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { For real-world problems, the task of decision-makers is to identify a solution that can satisfy a set of performance criteria, which are often in conflict with each other. Multi-objective evolutionary algorithms tend to focus on obtaining a family of solutions that represent the trade-offs between the criteria; however ultimately a single solution must be selected. This need has driven a requirement to incorporate decision-maker preference models into such algorithms, a technique that is very common in the wider field of multiple criteria decision making. This paper reviews techniques which have combined evolutionary multi-objective optimisation and multiple criteria decision making. Three classes of hybrid techniques are presented: a posteriori, a priori, and interactive, including methods used to model the decision-makers preferences and example algorithms for each category. To encourage future research directions, a commentary on the remaining issues within this research area is also provided. }} @InProceedings{Alhindi:2014:CEC, title = {{MOEA/D} with Tabu Search for Multiobjective Permutation Flow Shop Scheduling Problems}, author = {Ahmad Alhindi and Qingfu Zhang}, pages = {1155--1164}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms}, abstract = { Multiobjective Evolutionary Algorithm based on Decomposition (MOEA/D) decomposes a multiobjective optimisation problem into a number of single objective problem and optimises them in a collaborative manner. This paper investigates how to use Tabu Search (TS), a well-studied single objective heuristic to enhance MOEA/D performance. In our proposed approach, the TS applies to these subproblems with the aim to escape from local optimal solutions. The experimental studies have shown that MOEA/D with TS outperforms the classical MOEA/D on multiobjective permutation flow shop scheduling problems. It also have demonstrated that use of problem specific knowledge can significantly improve the algorithm performance. }} @InProceedings{Cheung:2014:CEC, title = {Online Objective Reduction for Many-Objective Optimization Problems}, author = {Yiu-ming Cheung and Fangqing Gu}, pages = {1165--1171}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms}, abstract = { For many-objective optimisation problems, i.e. the number of objectives is greater than three, the performance of most of the existing Evolutionary Multi-objective Optimisation algorithms will deteriorate to a certain degree. It is therefore desirable to reduce many objectives to fewer essential objectives, if applicable. Currently, most of the existing objective reduction methods are based on objective selection, whose computational process is, however, laborious. In this paper, we will propose an online objective reduction method based on objective extraction for the many-objective optimisation problems. It formulates the essential objective as a linear combination of the original objectives with the combination weights determined based on the correlations of each pair of the essential objectives. Subsequently, we will integrate it into NSGA-II. Numerical studies have show the efficacy of the proposed approach. }} @InProceedings{Gee:2014:CEC, title = {Diversity Preservation with Hybrid Recombination for Evolutionary Multiobjective Optimization}, author = {Sen Bong Gee and Kay Chen Tan}, pages = {1172--1178}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Multiobjective optimisation}, abstract = { Convergence and diversity are two crucial issues in evolutionary multiobjective optimisation. To enhance the diversity property of Multiobjective Evolutionary Algorithm (MOEA), a novel selection method is implemented on decomposition-based MOEA (MOEA/D). The selection method incorporates the concept of maximum diversity loss, which quantifies the diversity loss of each individual in every generation. By monitoring tolerance of the diversity loss, the diversity of the solutions in each generation can be preserved. To further enhance the algorithm's search ability, a new hybrid recombination strategy is implemented by taking the advantage of different recombination operator. In terms of Inverted Generational Distance (IGD), the experiment results shown that the proposed algorithm, namely DHRSMOEA/D, performed significantly better than many state-of-the-art MOEAs in most of the CEC-09 and WFG test problems. }} @InProceedings{Alicino:2014:CEC, title = {An Evolutionary Approach to the Solution of Multi-Objective Min-Max Problems in Evidence-Based Robust Optimization}, author = {Simone Alicino and Massimiliano Vasile}, pages = {1179--1186}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Multi-objective evolutionary algorithms, Bilevel Optimisation}, abstract = { This paper presents an evolutionary approach to solve the multi-objective min-max problem (MOMMP) that derives from the maximisation of the Belief in robust design optimisation. In evidence-based robust optimisation, the solutions that minimise the design budgets are robust under epistemic uncertainty if they maximise the Belief in the realisation of the value of the design budgets. Thus robust solutions are found by minimising, with respect to the design variables, the global maximum with respect to the uncertain variables. This paper presents an algorithm to solve MOMMP, and a computational cost reduction technique based on Kriging meta models. The results show that the algorithm is able to accurately approximate the Pareto front for a MOMMP at a fraction of the computational cost of an exact calculation. }} @InProceedings{Luo:2014:CEC, title = {Kriging Model Based Many-Objective Optimization with Efficient Calculation of Expected Hypervolume Improvement}, author = {Chang Luo and Koji Shimoyama and Shigeru Obayashi}, pages = {1187--1194}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Meta-modelling and surrogate models, Numerical optimisation}, abstract = { The many-objective optimisation performance of using expected hypervolume improvement (EHVI) as the updating criterion of Kriging surrogate model is investigated, and compared with those of using expected improvement (EI) and estimation (EST) updating criteria in this paper. An exact algorithm to calculate hypervolume is used for the problems with less than six objectives. On the other hand, in order to improve the efficiency of hypervolume calculation, an approximate algorithm to calculate hypervolume based on Monte Carlo sampling is adopted for the problems with more objectives. Numerical experiments are conducted in 3 to 12-objective DTLZ1, DTLZ2, DTLZ3 and DTLZ4 problems. The results show that, in DTLZ3 problem, EHVI always obtains better convergence and diversity performances than EI and EST for any number of objectives. In DTLZ2 and DTLZ4 problems, the advantage of EHVI is shown gradually as the number of objectives increases. The present results suggest that EHVI will be a highly competitive updating criterion for the many-objective optimisation with Kriging model. }} % Session: WeE1-2 Evolutionary Games and Multi-Agent Systems @InProceedings{Sudo:2014:CEC, title = {Effects of Ensemble Action Selection on the Evolution of Iterated Prisoner's Dilemma Game Strategies}, author = {Takahiko Sudo and Yusuke Nojima and Hisao Ishibuchi}, pages = {1195--1201}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games, Games}, abstract = { Iterated prisoner's dilemma (IPD) games have been frequently used for examining the evolution of cooperative game strategies. It has been pointed out in some studies that the choice of a representation scheme (i.e., coding mechanism) has a large effect on the evolution. A choice of a different representation scheme often leads to totally different results. In those studies on IPD games, a single representation scheme is assigned to all players. That is, all players have the same representation scheme. In our former studies, we reported experimental results in an inhomogeneous setting where a different representation scheme was assigned to each player. The evolution of cooperation among different types of game strategies was examined. In this paper, we report experimental results in another interesting setting where each player is assumed to have multiple strategies with different representation schemes. The next action of each player is determined by a majority vote by its strategies. That is, each player is assumed to have an ensemble decision making system. Experimental results in such an ensemble IPD model are compared with those in the standard IPD model where each player has a single strategy. }} @InProceedings{Tsang:2014:CEC, title = {The Structure of a Probabilistic 2-State Finite Transducer Representation for Prisoner's Dilemma}, author = {Jeffrey Tsang}, pages = {1202--1209}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games}, abstract = { Several studies have used the fingerprint, a mathematical technique that generates a representation-independent functional signature of a game playing strategy, to conduct automated analyses of spaces of strategies. This study looks at an even larger state space, namely a grid over the probabilistic 2-state finite transducers, as a representation for playing Prisoner's Dilemma. Even using just a three-level \{0, 0.5, 1\} grid amounts to 100,000 representable strategies, with an immense 40,679 unique strategies within. All strategies are fingerprinted and all pairwise distances computed, then hierarchical clustering reduces this dataset to around size 10,000 for further analysis with multidimensional scaling. Results indicate that the 20-dimensional grid has no obvious cutoff scales of structure, that we can quantify several important dimensions, and a high level of similarity with past results on smaller state spaces. We also find an interesting difference between complete playing equivalence of deterministic versus probabilistic transducers. }} @InProceedings{Scheepers:2014:CEC, title = {Competitive Coevolutionary Training of Simple Soccer Agents from Zero Knowledge}, author = {Christiaan Scheepers and Andries Engelbrecht}, pages = {1210--1217}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary games and multi-agent systems, Particle swarm optimisation (PSO), Computational Intelligence and Games}, abstract = { A new competitive coevolutionary team-based particle swarm optimisation (CCPSO) algorithm is developed to train multi-agent teams from zero knowledge. The CCPSO algorithm uses the charged particle swarm optimiser to train neural network controllers for simple soccer agents. The training performance of the CCPSO algorithm is analysed. The analysis identifies a critical weakness of the CCPSO algorithm in the form of outliers in the measured performance of the trained players. A hypothesis is presented that explains the presence of the outliers, followed by a detailed discussion of various biased and unbiased relative fitness functions. A new relative fitness function based on FIFA's league ranking system is presented. The performance of the unbiased relative fitness functions is evaluated and discussed. The final results show that the FIFA league ranking relative fitness function outperforms the other unbiased relative fitness functions, leading to consistent training results. }} @InProceedings{Greenwood:2014:CEC, title = {Online Generation of Trajectories for Autonomous Vehicles Using a Multi-Agent System}, author = {Garrison Greenwood and Saber Elsayed and Ruhul Sarker and Hussein Abbass}, pages = {1218--1224}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary games and multi-agent systems, Adaptive dynamic programming and reinforcement learning, Intelligent systems applications}, abstract = { Autonomous vehicles are frequently deployed in environments where only certain trajectories are feasible. Classical trajectory generation methods attempt to find a feasible trajectory that satisfies a set of constraints. In some cases the optimal trajectory may be known, but it is hidden from the autonomous vehicle. Under such circumstance the vehicle must discover a feasible trajectory. This paper describes a multi-agent system that uses a combination of reinforcement learning and differential evolution to generate a trajectory that is epsilon-close to a target trajectory that is hidden. }} @InProceedings{Lee:2014:CECa, title = {A Cooperative Coevolutionary Approach to Multi-Robot Formation Control}, author = {Seung-Mok Lee and Hyun Myung}, pages = {1225--1231}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Robotics, Evolutionary games and multi-agent systems, Coevolutionary systems}, abstract = { This paper proposes a cooperative coevolutionary approach to multi-robot formation control. To deal with the formation control problem, the concept of a cooperative coevolution (CC) framework is incorporated with model predictive control (MPC) such that candidates of all robots coevolve toward a Nash equilibrium in a distributed way. Using the Nash-equilibrium strategy, the robots can quickly move to a desired formation from their initial locations. The stability is guaranteed via a novel repair algorithm that enforces each candidate to satisfy a derived condition for asymptotic stability. The cooperative coevolutionary particle swarm optimisation (CCPSO) is adopted and modified to fit into the formation control problem. Simulations are performed on a group of nonholonomic mobile robots to demonstrate the effectiveness of the CC-based MPC. Also, the proposed MPC shows a better performance compared to sequential quadratic programming (SQP)-based MPC. }} @InProceedings{Li:2014:CECe, title = {Graph Centrality Measures and the Robustness of Cooperation}, author = {Menglin Li and Colm O'Riordan}, pages = {1232--1237}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary games and multi-agent systems}, abstract = { Previous research shows that for structured populations located on a graph, one of the most important attributes that can decide whether a cooperative community is robust is the topology of the graph. However, even in a graph which is highly robust with respect to cooperation, ``weak points'' may still exist which will allow the defection to spread quickly in the community. Previous work shows that the transitivity and the average degree are related to the robustness of the cooperation in the entire graph. In addition to considering the cooperation across the entire graph, whether an individual in the graph will allow the spread of defection is an important research question in its own right. In this work, we are trying to identify both the ``weak'' individuals and the the ``robust'' ones. We measure the centrality in the graph together with the the degree, the local clustering coefficient, the betweenness, the closeness, the degree eigenvector, and a few new designed centrality measures such as ``clustering eigenvector centrality''. The results show that for graphs that have a fixed number of vertices and edges, there are both robust individuals and weak individuals and that the higher the transitivity of the graph, the more robust the individuals are in the graph. However, although some of the graph centrality measures may indicate whether a vertex is robust or not, the prediction is still quite unstable. }} % Special Session: WeE1-3 Hybrid Evolutionary Computational Methods for Complex Optimisation Problems @InProceedings{Ling:2014:CEC, title = {Non-Invasive Detection of Hypoglycemic Episodes in Type1 Diabetes Using Intelligent Hybrid Rough Neural System}, author = {Sai Ho Ling and Phyo Phyo San and Hak Keung Lam and Hung Nguyen}, pages = {1238--1242}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Hybrid-evolutionary Computational Methods for Complex Optimisation Problems}, abstract = { Insulin-dependent diabetes mellitus is classified as Type 1 diabetes and it can be further classified as immune-mediated or idiopathic. Through the analysis of electrocardiographic (ECG) signals of 15 children with T1DM, an effective hypoglycemia detection system, hybrid rough set based neural network (RNN) is developed by the use of physiological parameters of ECG signal. In order to detect the status of hypoglycemia, the feature of ECG of type 1 diabetics are extracted and classified according to corresponding glucose levels. In this technique, the applied physiological inputs are partitioned into predicted (certain) or random (uncertain) parts using defined lower and boundary of rough regions. In this way, the neural network is designed to deal only with the boundary region which mainly consists of a random part of applied input signal causing inaccurate modelling of the data set. A global training algorithm, hybrid particle swarm optimisation with wavelet mutation (HPSOWM) is introduced for parameter optimisation of proposed RNN. The experiment is carried out using real data collected at Department of Health, Government of Western Australia. It indicated that the proposed hybrid architecture is efficient for hypoglycemia detection by achieving better sensitivity and specificity with less number of design parameters. }} @InProceedings{Chan:2014:CEC, title = {Image Deblurring Using a Hybrid Optimization Algorithm}, author = {Kit Yan Chan and N. Rajakaruna and C. Rathnayake and I. Murray}, pages = {1243--1249}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Hybrid-evolutionary Computational Methods for Complex Optimisation Problems, Real-world applications}, abstract = { In many applications, such as way finding and navigation, the quality of image sequences are generally poor, as motion blur caused from body movement degrades image quality. It is difficult to remove the blurs without prior information about the camera motion. In this paper, we use inertial sensors, including accelerometers and gyroscopes, installed in smart phones, in order to determine geometric data of camera motion during exposure. Based on the geometric data, we derive a blurring function namely point spread function (PSF) which deblur the captured image by reversing motion effect. However, determination of the optimal PSF with respect to the image quality is multi-optimum, as deblurred images are not linearly correlated to image intelligibility. Therefore, this paper proposes a hybrid optimisation method, which is, incorporated the mechanisms of particle swarm optimisation (PSO) and gradient search method, in order to optimise PSF parameters. It aims to incorporate the advantages of the two methods, where the PSO is effective in localising the global region and the gradient search method is effective in converging local optimum. Experimental results indicated that deblurring can be successfully performed using the optimal PSF. Also, the performance of proposed method is compared with the commonly used deblurring methods. Better results in term of image quality can be achieved. The resulting deblurring methodology is an important component. It will be used to improve deblurred images to perform edge detection, in order to detect paths, stairs ways, movable and immovable objects for vision-impaired people. }} @InProceedings{Yuwono:2014:CEC, title = {An Algorithm for Scalable Clustering: Ensemble Rapid Centroid Estimation}, author = {Mitchell Yuwono and Steven W. Su and Bruce D. Moulton and Ying Guo and Hung T. Nguyen}, pages = {1250--1257}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Hybrid-evolutionary Computational Methods for Complex Optimisation Problems, Particle swarm optimisation (PSO), Classification, clustering and data analysis}, abstract = { This paper describes a new algorithm, called Ensemble Rapid Centroid Estimation (ERCE), designed to handle large-scale non-convex cluster optimisation tasks, and estimate the number of clusters with quasi-linear complexity. ERCE stems from a recently developed Rapid Centroid Estimation (RCE) algorithm. RCE was originally developed as a lightweight simplification of the Particle Swarm Clustering (PSC) algorithm. RCE retained the quality of PSC, greatly reduced the computational complexity, and increased the stability. However, RCE has certain limitations with respect to complexity, and is unsuitable for non-convex clusters. The new ERCE algorithm presented here addresses these limitations. }} @InProceedings{Yu:2014:CECd, title = {Evolutionary Regional Network Modeling for Efficient Engineering Optimization}, author = {Jyh-Cheng Yu and Zhi-Fu Liang}, pages = {1258--1264}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Hybrid-evolutionary Computational Methods for Complex Optimisation Problems, Evolved neural networks, Engineering applications}, abstract = { This study presents a soft computing based optimisation methodology, the evolutionary regional neural network modelling for engineering applications with sampling constraints. Engineering optimisation often involves expensive experiment costs. Intelligent optimisation advocates establishing a neural network model using small training samples such as orthogonal array to set up a surrogate model for the engineering system followed by an optimum search in the model to reduce optimisation cost. However, scarce training samples might compromise modelling generality for a complex problem. Empirical rules suggest reliable predictions are likely restricted to the neigh boring space of training samples, and interpolating designs are more reliable than extrapolating designs. To avoid imperfection of model due to small learning samples, an evolutionary regional network model is set up to confine the search of quasi-optimum using genetic algorithm. The constrained search in the regional network model provides a reliable quasi-optimum. The verification of the optimum is added to the learning samples to retrain the regional network model while the size and the distribution of reliable space will evolve intelligently during the optimisation iteration using a fuzzy inference according to the prediction accuracy. An engineering case study, the optimal die gap parison programming of extrusion blow molding process for a uniform thickness, is presented to demonstrate the robustness and efficiency of the proposed methodology. }} @InProceedings{Li:2014:CECf, title = {Quantum Bacterial Foraging Optimization Algorithm}, author = {Fei Li and Yuting Zhang and Haibo Li}, pages = {1265--1272}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Quantum Computing and Evolutionary Computation, Discrete and combinatorial optimisation, Molecular and quantum computing}, abstract = { This paper proposes a novel swarm intelligence optimisation method which integrates bacterial foraging optimisation (BFO) with quantum computing, called quantum bacterial foraging optimisation (QBFO) algorithm. In QBFO, a multi-qubit which can represent a linear superposition of states in search space probabilistically is used to represent a bacterium, so that the quantum bacteria representation has a better characteristic of population diversity. A quantum rotation gate is designed to simulate the chemotactic step to drive the bacteria toward better solutions. Several tests are conducted based on benchmark functions including multi-peak function to evaluate optimisation performance of the proposed algorithm. The numeric results show that the proposed QBFO has more powerful properties in convergence rate, stability and the ability of searching for the global optimal solution than the original BFO and quantum genetic algorithm. In addition, we applied our proposed QBFO to solve the travellings salesman problem, which is a well-known NP-hard problem in combinatorial optimisation. The results indicate that the proposed QBFO shows better convergence behaviour without premature convergence, and has more powerful properties in convergence rate, stability and the ability of searching for the global optimal solution, as compared to ant colony optimisation algorithm and quantum genetic algorithm. }} @InProceedings{Liu:2014:CECi, title = {A Cultural Algorithm for Spatial Forest Harvest Scheduling}, author = {Wan-Yu Liu and Chun-Cheng Lin}, pages = {1273--1276}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications}, abstract = { This paper proposes a cultural algorithm for the spatial forest harvest scheduling for maximising the total harvested timber volume, under the constraints of minimum harvest age, minimum adjacency green-up age, and approximately even volume flow for each period of the schedule. In order to increase the solution-search ability, the cultural algorithm extracts problem specific information during the evolutionary solution search to update the belief space of a generation, which has cultural influences and guidance on the next generation. The key design of our cultural algorithm is to propose the cultural and evolutionary operators specifically for the problem. Experimental analysis shows that our cultural algorithm performs better than the previous approaches. }} % Special Session: WeE1-4 Large Scale Global Optimisation @InProceedings{Ye:2014:CEC, title = {A Hybrid Adaptive Coevolutionary Differential Evolution Algorithm for Large-Scale Optimization}, author = {Sishi Ye and Guangming Dai and Lei Peng}, pages = {1277--1284}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large Scale Global Optimisation, Differential evolution, Large-scale problems}, abstract = { In this paper, we propose a new algorithm, named HACC-D, for large scale optimisation problems. The motivation is to improve the optimisation method for the subcomponents in the cooperative coevolution framework. In the new HACC-D algorithm, an algorithm selection method named hybrid adaptive optimisation strategy is used. It is aimed to hybridise the superiority of two very efficient differential evolution algorithms, JADE and SaNSDE, as the subcomponent optimisation algorithm of the cooperative coevolution. In the beginning stage, the novel strategy evolves the initial population with JADE and SaNSDE as the subcomponent optimisation algorithm for a certain number of iterations separately. Then the one obtained better fitness value will be chosen to be the subcomponent optimisation algorithm for the following evolution process. In the later stage of evolution, the selected algorithm may be trapped in a local optimum or lose its ability to make further progress. So it exchanges the subcomponent optimisation algorithm with the other one when there is no improvement in the fitness every certain number of iterations. The proposed HACC-D algorithm is evaluated on CEC'2010 benchmark functions for large scale global optimisation. }} @InProceedings{Mahdavi:2014:CEC, title = {Cooperative Co-Evolution with a New Decomposition Method for Large-Scale Optimization}, author = {Sedigheh Mahdavi and Mohammad Ebrahim Shiri and Shahryar Rahnamayan}, pages = {1285--1292}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large Scale Global Optimisation, Coevolutionary systems}, abstract = { Cooperative Co-evolutionary algorithms are effective approaches to solve large-scale optimisation problems. The crucial challenge in these methods is the design of a decomposition method which is able to detect interactions among variables. In this paper, we proposed a decomposition method based on High Dimensional Model Representation (HDMR) which extracts separable and nonseparable subcomponents for Cooperative Co-evolutionary algorithms. The entire decomposition procedure is conducted before applying the optimisation. The experimental results for D=1000 on twenty CEC-2010 benchmark functions show that the proposed method is promisingly efficient to solve large-scale optimisation problems. The proposed approach is compared with two other methods and discussed in details. }} @InProceedings{Wei:2014:CECb, title = {Variable Grouping Based Differential Evolution Using an Auxiliary Function for Large Scale Global Optimization}, author = {Fei Wei and Yuping Wang and Tingting Zong}, pages = {1293--1298}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large Scale Global Optimisation, Genetic algorithms, Multi-objective evolutionary algorithms}, abstract = { Evolutionary algorithms (EAs) are a kind of efficient and effective algorithms for global optimisation problems. However, their efficiency and effectiveness will be greatly reduced for large scale problems. To handle this issue, a variable grouping strategy is first designed, in which the variables with the interaction each other are classified into one group, while the variables without interaction are classified into different groups. Then, evolution can be conducted in these groups separately. In this way, a large scale problem can be decomposed into several small scale problems and this makes the problem solving much easier. Furthermore, an auxiliary function, which can help algorithm to escape from the current local optimal solution and find a better one, is designed and integrated into EA. Based on these, variable grouping based differential evolution (briefly, VGDE)using auxiliary function is proposed. At last, the simulations are made on the standard benchmark suite in CEC'2013, and VGDE is compared with several well performed algorithms. The results indicate the proposed algorithm VGDE is more efficient and effective. }} @InProceedings{Wang:2014:CECc, title = {Solving Dynamic Double-Row Layout Problem via an Improved Simulated Annealing Algorithm}, author = {Shengli Wang and Xingquan Zuo and Xinchao Zhao}, pages = {1299--1304}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large Scale Global Optimisation, Engineering applications}, abstract = { Double-row layout problem (DRLP) is a new problem proposed in 2010. Different from single or multi-row layout problems, DRLP needs to determine not only the sequence of machines on both rows but also the exact location of each machine. Aiming at the dynamic environment of product processing in practice, in this paper we study DRLP under dynamic environment and propose a dynamic double-row layout problem (DDRLP) where the material flows may change over time. A mixed-integer programming model is established for the DDRLP. An improved simulated annealing (ISA) algorithm is proposed to for this problem. To represent a feasible solution, a mixed coding scheme is suggested to express the sequence of facilities and the exact location of each facility. Five operators are devised to make the ISA able to effectively solve this problem. Experiment results show that the proposed algorithm is able to find the optimal solutions for small size problem instances and outperform an exact approach (CPLEX) under limited run time for large size instances. }} @InProceedings{Omidvar:2014:CEC, title = {Effective Decomposition of Large-Scale Separable Continuous Functions for Cooperative Co-Evolutionary Algorithms}, author = {Mohammad Nabi Omidvar and Yi Mei and Xiaodong Li}, pages = {1305--1312}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems}, abstract = { In this paper we investigate the performance of cooperative co-evolutionary (CC) algorithms on large-scale fully-separable continuous optimisation problems. We have shown that decomposition can have significant impact on the performance of CC algorithms. The empirical results show that the subcomponent size should be chosen small enough so that the subcomponent size is within the capacity of the subcomponent optimiser. In practice, determining the optimal size is difficult. Therefore, adaptive techniques are desired by practitioners. Here we propose an adaptive method, MLSoft, that uses widely-used techniques in reinforcement learning such as the value function method and softmax selection rule to adapt the subcomponent size during the optimisation process. The experimental results show that MLSoft is significantly better than an existing adaptive algorithm called MLCC on a set of large-scale fully-separable problems. }} @InProceedings{Mei:2014:CEC, title = {Variable Neighborhood Decomposition for Large Scale Capacitated Arc Routing Problem}, author = {Yi Mei and Xiaodong Li and Xin Yao}, pages = {1313--1320}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems, Real-world applications}, abstract = { In this paper, a Variable Neighbourhood Decomposition (VND) is proposed for Large Scale Capacitated Arc Routing Problems (LSCARP). The VND employs the Route Distance Grouping (RDG) scheme, which is a competitive decomposition scheme for LSCARP, and generates different neighbourhood structures with different trade offs between exploration and exploitation. The search first uses a neighbourhood structure that is considered to be the most promising, and then broadens the neighbourhood gradually as it is getting stuck in a local optimum. The experimental studies show that the VND performed better than the state-of-the-art RDG-MAENS counterpart, and the improvement is more significant when the subcomponent size is smaller. This implies a great potential of combining the VND with small subcomponents. }} % Plenary Poster Session: PE3 Poster Session III @InProceedings{Ni:2014:CEC, title = {A New Dynamic Probabilistic Particle Swarm Optimization with Dynamic Random Population Topology}, author = {Qingjian Ni and Cen Cao and Xushan Yin}, pages = {1321--1327}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO)}, abstract = { Population topologies of Particle Swarm Optimisation algorithm (PSO) have direct impacts on the information sharing among particles during the evolution, and will influence the PSO algorithms' performance obviously. The canonical PSO algorithms usually use static population topologies, and the majority are the classic population topologies (such as fully connected topology and ring topology). In this paper, we present the strategies of dynamic random topology based on the random generation of population topologies. The basic idea is as follows: various random topologies are used at different stages of evolution in the population, and the solving performance of PSO algorithms is enhanced by improving the information exchange of population in different evolutionary stages. This provides a new way of thinking for the improvement of the PSO algorithm. Experimental results on a relatively new variant of dynamic probabilistic particle swarm optimisation show that our strategies can achieve better performance compared with traditional static population topologies. Experimental data are analysed and discussed in the paper, and the useful conclusions will provide a basis for further research. }} @InProceedings{Gu:2014:CEC, title = {An Adaptive {PSO} Based on Motivation Mechanism and Acceleration Restraint Operator}, author = {Jiangshao Gu and Xuanhua Shi}, pages = {1328--1336}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Heuristics, metaheuristics and hyper-heuristics, Self-adaptation in evolutionary computation}, abstract = { To obtain precise solutions in optimisation problems and decrease the risk of being trapped in local optima, researchers have studied on various improved particle swarm optimisations (PSO) and made a series of achievements. However, these methods focus on artificially altering the physical rules of motion, rather than strengthening the individual self-learning and adjustment during the optimisation process, which is the original motive of the swarm-based evolutionary algorithms. In this paper, we propose a fresh self-adaptive variant, MMARO-PSO, which employs motivation mechanism to simulate the behaviour of intelligent organisms more vividly. We manage to simplify the update formulae and give each term a definite bio-psychic sense. Furthermore, we introduce a vectorised operator to restrain particle's acceleration, instead of the inertia weight parameter in conventional methods. Large number of experiments were conducted and the results illustrate that these innovations make the technique perform more consistently to find a better balance between global exploration and local exploitation, compared with the existing versions, e.g. SPSO, e1-PSO, ARFPSO, and (k,l)PSO. }} @InProceedings{Zhang:2014:CECe, title = {The Enhanced Vector of Convergence for Particle Swarm Optimization Based on Constrict Factor}, author = {Wei Zhang and Yanan Gao and Chengxing Zhang}, pages = {1337--1342}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO)}, abstract = { The Particle Swarm Optimiser is used very widely for unimodal and multi-modal optimisation problems. Recently, most of variant PSOs are combing several evolutionary strategies in order to achieve a better performance on Benchmark functions, and even for shifted, rotated, or composite functions. In this paper, a new method known as Enhanced Vector of Convergence is proposed and combined with constrict factor to improve the convergence performance of Particle Swarm Optimiser. In experimental study, other 5 variant Particle Swarm Optimisers are compared, and acceptance rate, t-Test are used for further evaluation. The results indicate that the Enhance Vector of Convergence can significantly improve the accurate level of Particle Swarm Optimiser. }} @InProceedings{Xu:2014:CEC, title = {Evolutionary Semi-Supervised Learning with Swarm Intelligence}, author = {Xiaohua Xu and Lin Lu and Ping He and Jie Ding and Yongsheng Ju}, pages = {1343--1350}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation, Classification, clustering and data analysis}, abstract = { To address the issue of evolutionary data classification, we propose an evolving swarm classification model. It treats each class as an ant colony carrying different type of pheromone. The ant colonies send their members to propagate their unique pheromone on the unlabelled instances, so as to label them for member recruitment. Meanwhile, the unlabelled instances are treated as unlabelled ants, which also have their preferences for joining one of those label ed colonies. We call it homing feedback, and integrate it into the pheromone update process. Afterwards, the natural selection process is carried out to keep a balance between the member recruitment and the ant colony size maintenance. Sufficient experiments demonstrate that our algorithm is effective in the real-world evolutionary classification applications. }} @InProceedings{Zhang:2014:CECf, title = {A Fast Restarting Particle Swarm Optimizer}, author = {Junqi Zhang and Xiong Zhu and Wei Wang and Jing Yao}, pages = {1351--1358}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Numerical optimisation}, abstract = { Particle swarm optimisation (PSO) is a swarm intelligence technique that optimises a problem by iterative exploration and exploitation in the search space. However, PSO cannot achieve the preservation of population diversity on solving multi-model problems, and once the swarm falls into local convergence, it cannot jumps out of the local trap. In order to solve this problem, this paper presents a fast restarting particle swarm optimisation (FRPSO), which uses a novel restarting strategy based on a discrete finite-time particle swarm optimisation (DFPSO). Taking advantage of frequently speeding up the swarm to converge along with a greater exploitation capability and then jumping out of the trap, this algorithm can preserve population diversity and provide a superior solution. The experiment performs on twenty-five benchmark functions which consist of single-model, multi-model and hybrid composition problems, and the experimental result demonstrates that the performance of the proposed FRPSO algorithm is better than the other three representatives of the advanced PSO algorithm on most of these functions. }} @InProceedings{Li:2014:CECg, title = {Dimensions Cooperate by {Euclidean} Metric in Particle Swarm Optimization}, author = {Zezhou Li and Junqi Zhang and Wei Wang and Jing Yao}, pages = {1359--1366}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Coevolution and collective behaviour, Numerical optimisation}, abstract = { Since Particle Swarm Optimisation (PSO) was introduced, variants of PSO have usually updated velocities of particles in each dimension independently in the high-dimensional space. This paper proposes a Dimensionally Cooperative PSO (DCPSO), in which dimensions cooperate to update velocities of particles through Euclidean metric. The Euclidean metric first builds pbest-centered and gbest-centred hyper spheres. And then, velocity vectors of particles are derived from stochastic points obeying a distribution within the hyper spheres for dimensions cooperating. To the best knowledge of the authors, DCPSO is the first to investigate such cooperation of dimensions through Euclidean metric, instead of updating each dimension independently. Compared with the traditional PSO, DCPSO is validated by simulations on the 20 standard benchmark problems from CEC 2013. Furthermore, the differences between the behaviours of the traditional PSO and the proposed DCPSO are analysed from the aspect of the search space. Meanwhile, the curse of dimensionality is illustrated by comparisons between the traditional PSO and DCPSO in distinct dimensions. }} @InProceedings{Li:2014:CECh, title = {Biclustering of Gene Expression Data Using Particle Swarm Optimization Integrated with Pattern-Driven Local Search}, author = {Yangyang Li and Xiaolong Tian and Licheng Jiao and Xiangrong Zhang}, pages = {1367--1373}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation, Classification, clustering and data analysis, Particle swarm optimisation (PSO)}, abstract = { Biclustering is of great significance in the analysis of gene expression data and is proved to be a NP-hard problem. Among the existing intelligent optimisation algorithms used in the gene expression data analysis, most concentrate on the global search ability but ignore the inherent trajectory information of gene expression data, so the search efficiency is low. In this paper, a pattern-driven local search operator is incorporated in the binary Particle Swarm Optimisation (PSO) algorithm in order to improve the search efficiency. Experiments show that our approach is valid. }} @InProceedings{Shuai:2014:CEC, title = {Simulating the Coevolution of Language and Long-Term Memory}, author = {Lan Shuai and Zhen Wang and Tao Gong}, pages = {1374--1381}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Coevolutionary systems, Coevolution and collective behaviour, Evolutionary games and multi-agent systems}, abstract = { Memory is fundamental to social activities such as language communications, yet it remains unclear how memory capacity and language use influence each other during language evolution, especially the early stage of language origin. Here, we proposed an evolutionary framework to address this issue. It assumed a genetic transmission of memory capacity and integrated natural and cultural selections that respectively affected the choices of parents for reproducing offspring and teaching these offspring. Simulation results obtained under this framework and relevant statistical analyses collectively traced a coevolution of language and capacity of individual long-term memory for storing acquired linguistic knowledge during the origin of a communal language in a population. In line with the coevolutionary theory of language and related cognitive competences, this simulation study demonstrated that culturally-constituted aspects (communicative success) could drive the natural selection of predisposed cognitive features (long-term memory capacity), thus showing that language resulted from biological evolution, individual learning, and socio-cultural transmissions. }} @InProceedings{Chen:2014:CECb, title = {Evolutionary Clustering with Differential Evolution}, author = {Gang Chen and Wenjian Luo and Tao Zhu}, pages = {1382--1389}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Data mining}, abstract = { Evolutionary clustering is a hot research topic that clusters the time-stamped data and it is essential to some important applications such as data streams clustering and social network analysis. An evolutionary clustering should accurately reflect the current data at any time step while simultaneously not deviate too drastically from the recent past. In this paper, the differential evolution (DE) is applied to deal with the evolutionary clustering problem. Comparing with the typical k-means, evolutionary clustering based on DE (deEC) could perform a global search in the solution space. Experimental results over synthetic and real-world data sets demonstrate that the deEC provides robust and adaptive solutions. }} @InProceedings{Ameerudden:2014:CEC, title = {Smart Hybrid Genetic Algorithms in the Bandwidth Optimization of a {PIFA} Antenna}, author = {Mohammad Riyad Ameerudden and Harry Rughooputh}, pages = {1390--1396}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms}, abstract = { With the exponential development of mobile communications and the miniaturisation of radio frequency transceivers, the need for small and low profile antennas at mobile frequencies is constantly growing. Therefore, new antennas should be developed to provide both larger bandwidth and small dimensions. This paper presents a smart optimisation technique using a hybridised Genetic Algorithms (GA) and comparison with more classical GA techniques. The hybridisation involves primarily a clustering mechanism coupled with the intelligence of the Binary String Fitness Characterisation (BSFC) technique. The optimisation engine is applied to the design of a Planar Inverted-F Antenna (PIFA) in order to achieve an optimal bandwidth performance in the 2 GHz band. During the optimisation process, the PIFA is modelled and evaluated using the finite-difference time domain (FDTD) method. }} @InProceedings{Chen:2014:CECc, title = {Evolutionary Many-Objective Optimization by {MO-NSGA-II} with Enhanced Mating Selection}, author = {Shao-Wen Chen and Tsung-Che Chiang}, pages = {1397--1404}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { Many-objective optimisation deals with problems with more than three objectives. The rapid growth of non-dominated solutions with the increase of the number of objectives weakens the search ability of Pareto-dominance-based multiobjective evolutionary algorithms. MO-NSGA-II strengthens its dominance based predecessor, NSGA-II, by guiding the search process with reference points. In this paper, we further improve MO-NSGA-II by enhancing its mating selection mechanism with a hierarchical selection and a neighbourhood concept based on the reference points. Experimental results confirm that the proposed ideas lead to better solution quality. }} @InProceedings{Luo:2014:CECa, title = {A Niching Two-Layered Differential Evolution with Self-Adaptive Control Parameters}, author = {Yongxin Luo and Sheng Huang and Jinglu Hu}, pages = {1405--1412}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Self-adaptation in evolutionary computation}, abstract = { Differential evolution (DE) is an effective and efficient evolutionary algorithm in continuous space. The setting of control parameters is highly relevant with the convergence efficiency, and varies with different optimisation problems even at different stages of evolution. Self-adapting control parameters for finding global optima is a long-term target in evolutionary field. This paper proposes a two-layered DE (TLDE) with self adaptive control parameters combined with niching method based mutation strategy. The TLDE consists of two DE layers: a bottom DE layer for the basic evolution procedure, and a top DE layer for control parameter adaptation. Both layers follow the procedure of DE. Moreover, to mitigate the common phenomenon of premature convergence in DE, a clearing niching method is brought out in finding efficient mutation individuals to maintain diversity during the evolution and stabilise the evolution system. The performance is validated by a comprehensive set of twenty benchmark functions in parameter optimisation and competitive results are presented. }} @InProceedings{Lattarulo:2014:CEC, title = {Application of the {MOAA} for the Optimization of {CORAIL} Assemblies for Nuclear Reactors}, author = {Valerio Lattarulo and Benjamin A. Lindley and Geoffrey T. Parks}, pages = {1413--1420}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Multi-objective evolutionary algorithms, Multiobjective optimisation}, abstract = { The Multi-objective Alliance Algorithm (MOAA), a recently introduced optimisation algorithm, is used for the optimisation of heterogeneous low-enriched uranium (LEU) + mixed-oxide fuel (MOX) assemblies for pressurised water reactors (PWRs). This is a constrained nuclear problem with two objectives and a mixed-integer solution space. The efficacy of the algorithm is demonstrated through comparisons with NSGAII for between 300 and 2000 function evaluations. Through the epsilon and hypervolume indicators and the Kruskal-Wallis statistical test, we show that the MOAA outperforms NSGA-II on this problem. The MOAA was also able to find a set of solutions that are better than the 'expert solution' for this problem. }} @InProceedings{Pop:2014:CEC, title = {A Hybrid Approach Based on Genetic Algorithms for Solving the Clustered Vehicle Routing Problem}, author = {Petrica Pop and Camelia Chira}, pages = {1421--1426}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Heuristics, metaheuristics and hyper-heuristics, Memetic, multi-meme and hybrid algorithms}, abstract = { In this paper, we describe a hybrid approach based on the use of genetic algorithms for solving the Clustered Vehicle Routing Problem, denoted by CluVRP. The problem studied in this work is a generalisation of the classical Vehicle Routing Problem (VRP) and is closely related to the Generalised Vehicle Routing Problem (GVRP). Along with the genetic algorithm, we consider a local-global approach to the problem that is reducing considerably the size of the solutions space. The obtained computational results point out that our algorithm is an appropriate method to explore the search space of this complex problem and leads to good solutions in a reasonable amount of time. }} @InProceedings{Montgomery:2014:CEC, title = {Identifying and Exploiting the Scale of a Search Space in Differential Evolution}, author = {James Montgomery and Stephen Chen and Yasser Gonzalez-Fernandez}, pages = {1427--1434}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential Evolution: Past, Present and Future, Numerical optimisation, Self-adaptation in evolutionary computation}, abstract = { Optimisation in multimodal landscapes involves two distinct tasks: identifying promising regions and location of the (local) optimum within each region. Progress towards the second task can interfere with the first by providing a misleading estimate of a region's value. Thresheld convergence is a generally applicable "meta"-heuristic designed to control an algorithm's rate of convergence and hence which mode of search it is using at a given time. Previous applications of thresheld convergence in differential evolution (DE) have shown considerable promise, but the question of which threshold values to use for a given (unknown) function landscape remains open. This work explores the use of clustering-based method to infer the distances between local optima in order to set a series of decreasing thresholds in a multi-start DE algorithm. Results indicate that on those problems where normal DE converges, the proposed strategy can lead to sizable improvements. }} @InProceedings{Ksibi:2014:CEC, title = {Enhancing Relevance Re-Ranking Using Nature-Inspired Meta-Heuristic Optimization Algorithms}, author = {Amel Ksibi and Anis Ben Ammar and Chokri Ben Amar}, pages = {1435--1442}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Convergence, scalability and complexity analysis, Evolutionary programming}, abstract = { Over the last years, Relevance re-ranking has been an attractive research, aiming to re-order the initial image search result list by which relevant results should be at the top ranking list and irrelevant results should be pruned. In this paper, we propose to explore two population-based meta-heuristic algorithms, which are Particle Swarm optimisation(PSO), and Cuckoo search(CS), in order to solve the relevance re-ranking problem as a constrained regularisation framework. By doing so, we define two reranking processes, referred to as APSORank and CS-Rank that converge to the optimal ranked list. Results are further provided to demonstrate the effectiveness and performance of these two reranking processes. }} @InProceedings{Kromer:2014:CEC, title = {Can Deterministic Chaos Improve Differential Evolution for the Linear Ordering Problem?}, author = {Pavel Kromer and Ivan Zelinka and Vaclav Snasel}, pages = {1443--1448}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Evolutionary Computing with Deterministic Chaos}, abstract = { Linear ordering problem is a popular NP-hard combinatorial optimisation problem attractive for its complexity, rich library of test data, and variety of real world applications. It has been solved by a number of heuristic as well as metaheuristic methods in the past. The implementation of nature-inspired metaheuristic optimisation and search methods usually depends on streams of integer and floating point numbers generated in course of their execution. The pseudo-random numbers are used for an in-silico emulation of probability-driven natural processes such as arbitrary modification of genetic information (mutation, crossover), partner selection, and survival of the fittest (selection, migration) and environmental effects (small random changes in particle motion direction and velocity). Deterministic chaos is a well known mathematical concept that can be used to generate sequences of seemingly random real numbers within selected interval in a predictable and well controllable way. In the past, it has been used as a basis for various pseudo-random number generators with interesting properties. Recently, it has been shown that it can be successfully used as a source of stochasticity for nature-inspired algorithms solving a continuous optimisation problem. In this work we compare effectiveness of the differential evolution with different pseudo-random number generators and chaotic systems as sources of stochasticity when solving the linear ordering problem. }} @InProceedings{Zhang:2014:CECg, title = {Two Parameter Update Schemes for Recurrent Reinforcement Learning}, author = {Jin Zhang and Dietmar Maringer}, pages = {1449--1453}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Finance and economics, Intelligent systems applications}, abstract = { Recurrent reinforcement learning (RRL) is a machine learning algorithm which has been proposed by researchers for constructing financial trading platforms. When an analysis of RRL trading performance is conducted using low frequency financial data (e.g. daily data), the weakening autocorrelation in price changes may lead to a decrease in trading profits as compared to its applications in high frequency trading. There therefore is a need to improve RRL for the purposes of daily equity trading. This paper presents two parameter update schemes (the `average elitist' and the `multiple elitist') for RRL. The purpose of the first scheme is to improve out-of-sample performance of RRL-type trading systems. The second scheme aims to exploit serial dependence in stock returns to improve trading performance, when traders deal with highly correlated stocks. Profitability and stability of the trading system are examined by using four groups of Standard and Poor stocks for the period January 2009 to December 2012. It is found that the Sharpe ratios of the stocks increase after we use the two parameter update schemes in the RRL trading system. }} @InProceedings{Li:2014:CECi, title = {Differential Evolution Strategy Based on the Constraint of Fitness Values Classification}, author = {Zhihui Li and Zhigang Shang and J. J. Liang and B. Y. Qu}, pages = {1454--1460}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { This paper presents a new Differential Evolution (DE) strategy, named as FCDE, based on the constraint of classification of fitness function values. To ensure the population could move to the better fitness landscape, the global fitness value distribution information of the objective function are used and all points in the population are classified into three class by their fitness values in each generation, so the points in each class choose their donor vector and differential vector from the points in adjacent senior class to form the trial vector. This strategy could speed up the convergence to global optimal as well as avoid falling into the local optimal. Another attractive character of FCDE is the control parameters in this DE variant are self-adaptive. This method is tested on the 30 benchmark functions of CEC2014 special session and competition on single objective real-parameter numerical optimisation. The experimental results showed acceptable reliability of this strategy in high search dimension. This paper will participate in the competition on real parameter single objective optimisation to compare with other algorithms. }} @InProceedings{Htiouech:2014:CEC, title = {A {Lagrangian} and Surrogate Information Enhanced Tabu Search for the {MMKP}}, author = {Skander Htiouech and Sadok Bouamama}, pages = {1461--1468}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Discrete and combinatorial optimisation}, abstract = { The multidimensional multi-choice knapsack problem (MMKP) is NP-hard. Within the framework of solving this problem, we suggest newer approaches. We not only propose a multi-starts version of our previous works approach using surrogate constraint information based choices [31][32], but also we introduce another newer heuristic. The latter uses Lagrangian relaxation information in place of surrogate information. Compared with other literature known methods described so far, our approaches results are competitive. }} @InProceedings{Yang:2014:CECa, title = {Estimation of Distribution Algorithms Based Unmanned Aerial Vehicle Path Planner Using a New Coordinate}, author = {Peng Yang and Ke Tang and Jose Antonio Lozano}, pages = {1469--1476}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Collaborative Learning and Optimisation}, abstract = { Path planning technique is vital to Unmanned Aerial Vehicle (UAV). Evolutionary Algorithms (EAs) have been widely used in planning path for UAV. In these EA-based path planners, Cartesian coordinate system and polar coordinate system are commonly used to codify the path. However, either of them has its drawback: Cartesian coordinate systems result in an enormous search space, whilst polar coordinate systems are unfit for local modifications resulting e.g., from mutation and/ or crossover. In order to overcome these two drawbacks, we solve the UAV path planning in a new coordinate system. As the new coordinate system is only a rotation of Cartesian coordinate system, it is inherently easy for local modification. Besides, this new coordinate system has successfully reduced the search space by explicitly dividing the mission space into several subspaces. Within this new coordinate system, an Estimation of Distribution Algorithms (EDAs) based path planner is proposed in this paper. Some experiments have been designed to test different aspects of the new path planner. The results show the effectiveness of this planner. }} @InProceedings{Wu:2014:CECb, title = {An Uncultivated Wolf Pack Algorithm for High-Dimensional Functions and Its Application in Parameters Optimization of {PID} Controller}, author = {Husheng Wu and Fengming Zhang and Lushan Wu}, pages = {1477--1482}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Artificial ecology and artificial life, Single Objective Numerical Optimisation}, abstract = { To solve high-dimensional function optimisation problems, many swarm intelligent algorithms have been proposed. Inspired by hunting behaviour and distribution mode of uncultivated and barbarous wolf pack, we proposed a method, named uncultivated wolf pack algorithm (UWPA). Experiments are conducted on a suit of high-dimensional benchmark functions with different characteristics. What's more, the compared simulation experiments with other three typical intelligent algorithms, show that UWPA has better convergence and robustness. At last, this algorithm is successfully applied in parameters searching for PID controller. }} @InProceedings{Marchetti:2014:CEC, title = {On the Inference of Deterministic Chaos: Evolutionary Algorithm and Metabolic {P} System Approaches}, author = {Luca Marchetti and Vincenzo Manca and Ivan Zelinka}, pages = {1483--1489}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Emergent technologies, Evolutionary Computing with Deterministic Chaos}, abstract = { This paper shows the possibility of using Metabolic P systems (MP systems) for chaotic system identification-reconstruction and it compares presented results with previous ones obtained by evolutionary algorithms. An important potentiality of MP theory is given by its powerful computational chaos generation that can be also used as an internal module of evolutionary algorithms by increasing their ability in specific cases of their application. Reported numerical experiments are discussed at the end. }} @InProceedings{Yang:2014:CECb, title = {A New Method and Application for Controlling the Steady-State Probability Distributions of Probabilistic {Boolean} Networks}, author = {Meng Yang and Rui Li and Tianguang Chu}, pages = {1490--1495}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Biometrics, bioinformatics and biomedical applications, Multiobjective optimisation, Genetic algorithms}, abstract = { Probabilistic Boolean networks (PBNs) have been proved to be a useful tool for modelling genetic regulatory interactions. The study of the steady-state probability distribution may help to understand the essential long-run behaviour of a PBN. In this paper we focus on a type of PBNs derived from gene expression data collected in a study of metastatic melanoma. The metastatic melanoma model is usually described by a PBN containing seven genes among which WNT5A plays a significant role in the development of melanoma and is known to induce the metastasis of melanoma when highly active. This paper investigates the issue of how to drive the corresponding PBN towards desired steady-state probability distributions so as to reduce the WNT5A's ability to induce a metastatic phenotype. }} @InProceedings{He:2014:CECa, title = {Evolutionary Community Detection in Social Networks}, author = {Tiantian He and Keith C.C. Chan}, pages = {1496--1503}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Grouping and Graph-based Clustering}, abstract = { As people that share common characteristics and interests tend to communicate with each other more frequently, they form communities within social networks. Several methods have been developed to discover such communities based on topological metrics. These methods have been used to successfully discover communities that are relatively large, but for communities characterised by members interacting more frequently with each other rather than interacting with many others, we propose here an effective method which is based on the use of an evolutionary algorithm (EA) called ECDA. Given a social network represented as a graph, unlike existing approaches, ECDA considers both topological metrics of the graph and the attributes of the vertices and edges when detecting for communities in the network. It performs its task by formulating the community detection problem as an optimisation problem. By computing a measure of statistical significance for each attribute of the vertices, ECDA looks for communities in a network that have maximal connection significance within a community and minimal significance between any two communities. With such a strategy, ECDA partitions a network into different communities consisting of members with similar attributes within and different attributes without. Unlike other EAs, ECDA adopts a reproduction process consisting of special crossover and mutation operators, called Self-Evolution, to speed up the evolutionary process. ECDA has been tested with several real datasets and its performance is found to be very promising. }} @InProceedings{O'Neill:2014:CEC, title = {Experiments in Program Synthesis with Grammatical Evolution: A Focus on Integer Sorting}, author = {Michael O'Neill and Miguel Nicolau and Alexandros Agapitos}, pages = {1504--1511}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Grammatical Evolution, SBSE}, abstract = { We present the results of a series of investigations where we apply a form of grammar-based genetic programming to the problem of program synthesis in an attempt to evolve an Integer Sorting algorithm. The results confirm earlier research in the field on the difficulty of the problem given a primitive set of functions and terminals. The inclusion of a swap(i,j) function in combination with a nested for loop in the grammar enabled a successful solution to be found in every run. We suggest some future research directions to overcome the challenge of evolving sorting algorithms from primitive functions and terminals. }} @InProceedings{Pascoal:2014:CEC, title = {A Social-Evolutionary Approach to Compose a Similarity Function Used on Event Recommendation}, author = {Luiz Mario Lustosa Pascoal and Celso Goncalves Camilo-Junior and Edjalma Queiroz Silva and Thierson Couto Rosa}, pages = {1512--1519}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Real-world applications, Intelligent systems applications}, abstract = { With the development of web 2.0, social networks have achieved great space on the Internet, with that many users provide information and interests about themselves. There are expert systems that use the user's interests to recommend different products, these systems are known as Recommender Systems . One of the main techniques of a Recommender Systems is the Collaborative Filtering (User based) which recommends products to users based on what other similar people liked in the past. However, the methods to determine similarity between users have presented some problems. Therefore, this work presents a proposal of using social variables in the composition of the similarity function applied to a user on the recommendation of events. To test the proposal, details of friends and events of two target-users of the social network Facebook have been extracted. The results were compared with different deterministic heuristics, the Euclidean Distance and a aleatory method. The proposed model showed promising results and great potential to expand to different contexts. }} @InProceedings{Matei:2014:CEC, title = {Applying Evolutionary Computation for Evolving Ontologies}, author = {Oliviu Matei and Diana Contras and Petrica Pop}, pages = {1520--1527}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms}, abstract = { In this paper, we describe a novel application of evolutionary computation, namely for evolving ontologies. The general algorithm of evolutionary ontologies follow roughly the same guidelines as any other genetic algorithms. However, we introduced a new genetic operator, called repair, which is needed in order to make the offspring viable. Experiments for the generation of user centred automatically generated scenes demonstrate the performance of the proposed approach. }} % Special Session: WeE2-1 Evolutionary Computation in Dynamic and Uncertain Environments @InProceedings{Guo:2014:CEC, title = {Find Robust Solutions Over Time by Two-Layer Multi-Objective Optimization Method}, author = {Yinan Guo and Meirong Chen and Haobo Fu and Yun Liu}, pages = {1528--1535}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { Robust optimisation over time is a practical dynamic optimisation method, which provides two detailed computable metrics to get the possible robust solutions for dynamic scalar optimisation problems. However, the robust solutions fit for more time-varying moments or approximate the optimum more because only one metric is considered as the optimisation objective. To find the true robust solution set satisfying maximum both survival time and average fitness simultaneously during all dynamic environments, a novel two-layer multi-objective optimisation method is proposed. In the first layer, considering both metrics, the acceptable optimal solutions for each changing environment is found. Subsequently, they are composed of the practical robust solution set in the second layer. Taking the average fitness and the length of the robust solution set as two objectives, the optimal combinations for the whole time varying environments are explored. The experimental results for the modified moving peaks benchmark shows that the robust solution sets considering both metrics are superior to the robust solutions gotten by ROOT. As the key parameters, the fitness threshold has the more obvious impact on the performances of MROOT than the time window, whereas ROOT is more sensitive to both of them. }} @InProceedings{Hui:2014:CEC, title = {Niching-Based Self-adaptive Ensemble {DE} with {MMTS} for Solving Dynamic Optimization Problems}, author = {Sheldon Hui and Nagaratnam Suganthan Ponnuthurai}, pages = {1536--1541}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { Dynamic and non-stationary problems require optimisation algorithms search for the best solutions in a time-varying fitness environment. Various methods and strategies such as niching, clustering and sub-population approaches have been implemented with Differential Evolution (DE) to handle such problems. With the help of crowding niching to maintain general population diversity, this paper attempts to extend the Self-adaptive Ensemble DE with modified multi-trajectory search attempt to solve CEC2014 dynamic optimisation competition benchmark problems. }} @InProceedings{Mavrovouniotis:2014:CEC, title = {Interactive and Non-Interactive Hybrid Immigrants Schemes for Ant Algorithms in Dynamic Environments}, author = {Michalis Mavrovouniotis and Shengxiang Yang}, pages = {1542--1549}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments, Ant colony optimisation}, abstract = { Dynamic optimisation problems (DOPs) have been a major challenge for ant colony optimisation (ACO) algorithms. The integration of ACO algorithms with immigrants schemes showed promising results on different DOPs. Each type of immigrants scheme aims to address a DOP with specific characteristics. For example, random and elitism-based immigrants perform well on severely and slightly changing environments, respectively. In this paper, two hybrid immigrants, i.e., non-interactive and interactive, schemes are proposed to combine the merits of the aforementioned immigrants schemes. The experiments on a series of dynamic travelling salesman problems showed that the hybridisation of immigrants further improves the performance of ACO algorithms }} @InProceedings{Fu:2014:CECa, title = {What Are Dynamic Optimization Problems?}, author = {Haobo Fu and Peter Lewis and Bernhard Sendhoff and Ke Tang and Xin Yao}, pages = {1550--1557}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments}, abstract = { Dynamic Optimisation Problems (DOPs) have been widely studied using Evolutionary Algorithms (EAs). Yet, a clear and rigorous definition of DOPs is lacking in the Evolutionary Dynamic Optimisation (EDO) community. In this paper, we propose a unified definition of DOPs based on the idea of multiple-decision-making discussed in the Reinforcement Learning (RL) community. We draw a connection between EDO and RL by arguing that both of them are studying DOPs according to our definition of DOPs. We point out that existing EDO or RL research has been mainly focused on some types of DOPs. A conceptualised benchmark problem, which is aimed at the systematic study of various DOPs, is then developed. Some interesting experimental studies on the benchmark reveal that EDO and RL methods are specialised in certain types of DOPs and more importantly new algorithms for DOPs can be developed by combining the strength of both EDO and RL methods. }} @InProceedings{Chow:2014:CEC, title = {A Dynamic History-Driven Evolutionary Algorithm}, author = {Chi Kin Chow and Shiu Yin Yuen}, pages = {1558--1564}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Dynamic and uncertain environments}, abstract = { Dynamic objective problem (DOP) raises two challenging issues to evolutionary algorithm: comparing two individuals evaluated at different time instances and tracing the jumping global optimum. This paper presents a dynamic objective evolutionary algorithm (DOEA) that handles these issues through search history. The presented algorithm, namely dynamic objective history driven evolutionary algorithm(DyHdEA), stores the entire search history including the position, the fitness and the evaluated time of the solutions in a dynamic fitness tree. In the experiment section, DyHdEA is examined on a 10-dimensional DOP that is composed of five basis problems ranging from uni-modal to multi-modal, and from separable to non-separable. Meanwhile, the performance of DyHdEA is compared with five benchmark DOEAs including artificial immune algorithm, differential evolution, evolutionary programming, and particle swarm optimisation. Seen from the result, DyHdEA effectively traces the dynamic global optimum with jumping transitions. }} @InProceedings{Zhan:2014:CEC, title = {Adaptive Particle Swarm Optimization with Variable Relocation for Dynamic Optimization Problems}, author = {Zhi-Hui Zhan and Jun Zhang}, pages = {1565--1570}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Dynamic and uncertain environments, Self-adaptation in evolutionary computation}, abstract = { This paper proposes to solve the dynamic optimisation problem (DOP) by using an adaptive particle swarm optimisation (APSO) algorithm with an variable relocation strategy (VRS). The VRS based APSO algorithm (APSO/VRS) has the following two advantages when solving DOP. Firstly, by using the APSO optimising framework, the algorithm benefits from the fast optimization speed due to the adaptive parameter control. More importantly, the adaptive parameter and operator in APSO make the algorithm fast response to the environment changes of DOP. Secondly, VRS was reported in the literature to help dynamic evolutionary algorithm (DEA) to relocate the individual position in promising region when environment changes. Therefore, the modified VRS used in APSO can collect historical information in the stability stage and use such information to guide the particle variable relocation in the change stage. We evaluated both APSO and APSO/VRS on several dynamic benchmark problems and compared with two state-of-the-art DEAs and DEA that also used the VRS. The results show that both APSO and APSO/VRS can obtain very competitive results on these problems, and APSO/VRS outperforms others on most of the test cases. }} % Special Session: WeE2-2 Intelligent Design for Reliable Cloud Computing @InProceedings{Chang:2014:CEC, title = {Macroscopic Indeterminacy Swarm Optimization ({MISO}) Algorithm for Real-Parameter Search}, author = {Po-Chun Chang and Xiangjian He}, pages = {1571--1578}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent Design for Reliable Cloud Computing, Memetic, multi-meme and hybrid algorithms, Evolutionary programming}, abstract = { Swarm Intelligence (SI) is a nature-inspired emergent artificial intelligence. They are often inspired by the phenomena in nature. Many proposed algorithms are focused on designing new update mechanisms with formulae and equations to emerge new solutions. Despite the techniques used in an algorithm being the key factor of the whole system, the evaluation of candidate solutions also plays an important role. In this paper, the proposed algorithm Macroscopic Indeterminacy Swarm Optimisation (MISO) presents a new search scheme with indeterminate moment of evaluation. Here, we perform an experiment based on public benchmark functions. The results produced by MISO, Differential Evolution (DE) with various settings, Artificial Bee Colony (ABC), Simplified Swarm Optimisation (SSO), and Particle Swarm Optimisation (PSO) have been compared. The result shows MISO can achieve similar or even better performance than other algorithms. }} @InProceedings{Jiang:2014:CECb, title = {A Cooperative Honey Bee Mating Algorithm and Its Application in Multi-Threshold Image Segmentation}, author = {Yunzhi Jiang and Zhenlun Yang and Zhifeng Hao and Yinglong Wang and Huojiao He}, pages = {1579--1585}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent Design for Reliable Cloud Computing, Evolutionary programming, Heuristics, metaheuristics and hyper-heuristics}, abstract = { The problems of multi-threshold image segmentation remain great challenges for image compression, target recognition and computer vision. However, most of them are time-consuming. This paper proposes a cooperative honey bee mating-based algorithm (CHBMA) for image segmentation to save computation time while conquer the curse of dimensionality. CHBMA, based on honey bee mating algorithms (HBMA) and the cooperative learning, greatly enhances the search capability of the algorithm. Moreover, we adopt a new population initialisation strategy to make the search more efficient, according to the characters of multilevel thresholding in an image arranged from a low gray level to a high one. Extensive experiments have shown that CHBMA can deliver more effective and efficient results to be applied in complex image processing such as automatic target recognition, compared with state-of-the-art population-based thresholding methods. }} @InProceedings{Chou:2014:CEC, title = {A {RFID} Network Design Methodology for Decision Problem in Health Care}, author = {Chun-Hua Chou and Huang Chia-Ling and Po-Chun Chang}, pages = {1586--1592}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent Design for Reliable Cloud Computing, Real-world applications, Evolutionary fuzzy systems}, abstract = { This research extends our previous work on decision makers with methodology to optimise the design of a strategy for constructing Radio Frequency Identification (RFID). RFID technology is an automatic identification system through radio frequency for transferring data. Before deploying RFID system, one of the challenging problems is RFID network planning (RNP). The RNP problem must be solved to operate the large-scale network of readers, and need to satisfy a set of requires, such as coverage rate, economic, interference. This paper extends our previous work using soft computing technique to find the optimal positions of RFID readers based on Simplified Swarm Optimisation (SSO) algorithm. Meanwhile, the fuzzy-ART and K-means models are applied to efficiently and effectively search better solutions }} @InProceedings{Shang-Chia:2014:CEC, title = {Pareto Simplified Swarm Optimization for Grid-Computing Reliability and Service Makspan in Grid-{RMS}}, author = {Wei Shang-Chia and Yeh Wei-Chang and Yen Tso-Jung}, pages = {1593--1600}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Intelligent Design for Reliable Cloud Computing, Multi-objective evolutionary algorithms, Coevolutionary systems}, abstract = { In a grid-computing service, Grid-RMS must generate suitable assignment combinations (execution blocks) for dependable service quality and satisfactory makespan (service time). In this paper, service reliability of a grid environment and makespan of a grid application are estimated via the universal generating function methodology and probability theory. Then, we represent a simplified swarm optimisation (SSO) with the Pareto-set cluster (PC) to search the best assignment combinations in a grid environment with star topology. In terms of the task partition and distribution for a grid application, we employ a Pareto-set cluster to guide particle evolution, an elitist strategy to promote solution quality, and a simplified update mechanism to enhance the multi-objective optimisation effectiveness. Finally, we assess the performance of the PC-SSO by the interactive trade off problem based on the analysis of four scenarios with respect to the bi-objective problem and given restrictions. }} @InProceedings{Xu:2014:CECa, title = {A New Grouping Genetic Algorithm for the MapReduce Placement Problem in Cloud Computing}, author = {Xiaoyong Xu and Maolin Tang}, pages = {1601--1608}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Real-world applications, SBSE}, abstract = { MapReduce is a computation model for processing large data sets in parallel on large clusters of machines, in a reliable, fault-tolerant manner. A MapReduce computation is broken down into a number of map tasks and reduce tasks, which are performed by so called mappers and reducers, respectively. The placement of the mappers and reducers on the machines directly affects the performance and cost of the MapReduce computation. From the computational point of view, the mappers/reducers placement problem is a generation of the classical bin packing problem, which is NP-complete. Thus, in this paper we propose a new grouping genetic algorithm for the mappers/reducers placement problem in cloud computing. Compared with the original one, our grouping genetic algorithm uses an innovative coding scheme and also eliminates the inversion operator which is an essential operator in the original grouping genetic algorithm. The new grouping genetic algorithm is evaluated by experiments and the experimental results show that it is much more efficient than four popular algorithms for the problem, including the original grouping genetic algorithm. }} @InProceedings{Mohd-Yusoh:2014:CEC, title = {Composite {SaaS} Scaling in Cloud Computing Using a Hybrid Genetic Algorithm}, author = {Zeratul Mohd Yusoh and Maolin Tang}, pages = {1609--1616}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Real-world applications, SBSE}, abstract = { A Software-as-a-Service or SaaS can be delivered in a composite form, consisting of a set of application and data components that work together to deliver higher-level functional software. Components in a composite SaaS may need to be scaled, replicated or deleted, to accommodate the user's load. It may not be necessary to replicate all components of the SaaS, as some components can be shared by other instances. On the other hand, when the load is low, some of the instances may need to be deleted to avoid resource under use. Thus, it is important to determine which components are to be scaled such that the performance of the SaaS is still maintained. Extensive research on the SaaS resource management in Cloud has not yet addressed the challenges of scaling process for composite SaaS. Therefore, a hybrid genetic algorithm is proposed in which it utilises the problem's knowledge and explores the best combination of scaling plan for the components. Experimental results demonstrate that the proposed algorithm outperforms existing heuristic-based solutions. }} % Special Session: WeE2-3 Single Objective Numerical Optimisation II @InProceedings{Xu:2014:CECb, title = {A Differential Evolution with Replacement Strategy for Real-Parameter Numerical Optimization}, author = {Changjian Xu and Han Huang and ShuJin Ye}, pages = {1617--1624}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { Differential Evolution (DE) has been widely used as a continuous optimisation technique for several issues like electromagnetic optimisation, bioprocess system optimisation and so on. However, during the optimisation process, DE's population may stagnate local optimums and it may waste a large number of function evaluations for the population to get rid of them. This paper presents an improved DE algorithm (denoted as RSDE) which combines a Replacement Strategy (RS). The motivation of RS is that replacing an unimproved individual and replacing a premature population using RS can enhance the DE exploitation performance and exploration performance respectively. We tested the RSDE performance using the newly Single Objective Real-Parameter Numerical Optimisation problems provided by the CEC 2014 Special Session and Competition. Moreover, computational results and convergence figures are given for better compassion with other optimisation algorithm during the conference and afterwards. }} @InProceedings{Erlich:2014:CECa, title = {Evaluating the Mean-Variance Mapping Optimization on the {IEEE-CEC 2014} Test Suite}, author = {Istvan Erlich and Jose L. Rueda and Sebastian Wildenhues}, pages = {1625--1632}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { This paper provides a survey on the performance of the hybrid variant of the Mean-Variance Mapping Optimisation (MVMO-SH) when applied for solving the IEEE-CEC 2014 competition test suite on Single Objective Real-Parameter Numerical Optimisation. MVMO-SH adopts a swarm intelligence scheme, where each particle is characterised by its own solution archive and mapping function. Besides, multi-parent crossover is incorporated into the offspring creation stage in order to force the particles with worst fitness to explore other sub-regions of the search space. In addition, MVMO-SH can be customised to perform with an embedded local search strategy. Experimental results demonstrate the search ability of MVMO-SH for effectively tackling a variety of problems with different dimensions and mathematical properties }} @InProceedings{Molina:2014:CEC, title = {Influence of Regions on the Memetic Algorithm for the Special Session on Real-Parameter Single Objetive Optimisation}, author = {Daniel Molina and Benjamin Lacroix and Francisco Herrera}, pages = {1633--1640}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Memetic, multi-meme and hybrid algorithms, Numerical optimisation}, abstract = { Memetic algorithms with an appropriate trade-off between the exploration and exploitation can obtain very good results in continuous optimisation. That implies the evolutionary algorithm component should be focused in exploring the search space while the local search method exploits the achieved solutions. In a previous work, it was proposed a region-based algorithm, RMA-LSCh-CMA, adding to algorithm MA-LSCh-CMA a niching strategy that divides the domain search in equal hypercubes. The experimental results obtained, with the benchmark proposed in the CEC'2014 Special Session on Real-Parameter Single Objective Optimisation, show that the use of these regions allow the algorithm to obtain better results, specially in higher dimensions, and the resulting algorithm is more scalable. }} @InProceedings{Garden:2014:CEC, title = {Analysis and Classification of Optimisation Benchmark Functions and Benchmark Suites}, author = {Robert Garden and Andries Engelbrecht}, pages = {1641--1649}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Classification, clustering and data analysis}, abstract = { New and existing optimisation algorithms are often compared by evaluating their performance on a benchmark suite. This set of functions aims to evaluate the algorithm across a range of problems and serves as a baseline measurement of how the algorithm may perform on real-world problems. It is important that the functions serve as a good representative of commonly occurring problems. In order to select functions that will make up the benchmark suite, the characteristics and relationships among the functions must be known. This paper characterises the landscapes of two commonly used benchmark suites, and uses these landscape characteristics to obtain a high level view of the current state of benchmark functions. This is done by using a self-organising feature map to cluster and analyse functions based on landscape characteristics. It is found that while there are numerous functions that cover a wide range of characteristics, there are characteristics that are under represented, or not even covered at all. Furthermore, it is discovered that common benchmark suites are composed of functions which are highly similar according to the measured characteristics. }} @InProceedings{Elsayed:2014:CECb, title = {Testing United Multi-Operator Evolutionary Algorithms on the {CEC2014} Real-Parameter Numerical Optimization}, author = {Saber Elsayed and Ruhul Sarker and Daryl Essam and Noha Hamza}, pages = {1650--1657}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { This paper puts forward a proposal for combining multi-operator evolutionary algorithms (EAs), in which three EAs, each with multiple search operators, are used. During the evolution process, the algorithm gradually emphasises on the best performing multi-operator EA, as well as the search operator. The proposed algorithm is tested on the CEC2014 single objective real-parameter competition. The results show that the proposed algorithm has the ability to reach good solutions }} @InProceedings{Tanabe:2014:CEC, title = {Improving the Search Performance of {SHADE} Using Linear Population Size Reduction}, author = {Ryoji Tanabe and Alex Fukunaga}, pages = {1658--1665}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { SHADE is an adaptive DE which incorporates success-history based parameter adaptation and one of the state-of-the-art DE algorithms. This paper proposes L-SHADE, which further extends SHADE with Linear Population Size Reduction (LPSR), which continually decreases the population size according to a linear function. We evaluated the performance of L-SHADE on CEC2014 benchmarks and compared its search performance with state-of-the-art DE algorithms, as well as the state-of-the-art restart CMA-ES variants. The experimental results show that L-SHADE is quite competitive with state-of-the-art evolutionary algorithms. }} % Session: WeE2-4 Learning Classifier Systems @InProceedings{Karmaker-Santu:2014:CEC, title = {Towards Better Generalization in {Pittsburgh} Learning Classifier Systems}, author = {Shubhra Kanti Karmaker Santu and Md. Mustafizur Rahman and Md. Monirul Islam and Kazuyuki Murase}, pages = {1666--1673}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Learning classifier systems, Data mining, Genetic algorithms}, abstract = { Generalisation ability of a classifier is an important issue for any classification task. This paper proposes a new evolutionary system, i.e., EDARIC, based on the Pittsburgh approach for evolutionary machine learning and classification. The new system uses a destructive approach that starts with large-sized rules and gradually decreases the sizes as evolution progresses. Unlike most previous works, EDARIC adopts an intelligent deletion mechanism, evolves a separate population for each class of a given problem and uses an ensemble system to classify unknown instances. These features help in avoiding over-fitting and class-imbalance problems, which are beneficial for improving generalisation ability of a classification system. EDARIC also applies a rule post-processing step to exempt the evolution phase from the burden of tuning a large number of parameters. Experimental results on various benchmark classification problems reveal that EDARIC has better generalisation ability in case of both standard and imbalanced data-sets compared to many existing algorithms in the literature. }} @InProceedings{Scardapane:2014:CEC, title = {{GP}-Based Kernel Evolution for {L2}-Regularization Networks}, author = {Simone Scardapane and Danilo Comminiello and Michele Scarpiniti and Aurelio Uncini}, pages = {1674--1681}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Learning classifier systems, Classification, clustering and data analysis}, abstract = { In kernel-based learning methods, a crucial design parameter is given by the choice of the kernel function to be used. Although there is, in theory, an infinite range of potential candidates, a handful of kernels covers the majority of actual applications. Partly, this is due to the difficulty of choosing an optimal kernel function in absence of a-priori information. In this respect, Genetic Programming (GP) techniques have shown interesting capabilities of learning non-trivial kernel functions that outperform commonly used ones. However, experiments have been restricted to the use of Support Vector Machines (SVMs), and have not addressed some problems that are specific to GP implementations, such as diversity maintenance. In these respects, the aim of this paper is twofold. First, we present a customised GP-based kernel search method that we apply using an L2-Regularisation Network as the base learning algorithm. Second, we investigate the problem of diversity maintenance in the context of kernel evolution, and test an adaptive criterion for maintaining it in our algorithm. For the former point, experiments show a gain in accuracy for our method against fine-tuned standard kernels. For the latter, we show that diversity is decreasing critically fast during the GP iterations, but this decrease does not seems to affect performance of the algorithm. }} @InProceedings{Li:2014:CECj, title = {Generalized Classifier System: Evolving Classifiers with Cyclic Conditions}, author = {Xianneng Li and Wen He and Kotaro Hirasawa}, pages = {1682--1689}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Learning classifier systems, Adaptive dynamic programming and reinforcement learning}, abstract = { Accuracy-based XCS classifier system has been shown to evolve classifiers with accurate and maximally general characteristics. XCS generally represents its classifiers with binary conditions encoded in a ternary alphabet, i.e., \{0,1,hashtag\}, where hashtag is a ``don't care" symbol, which can match with 0 and 1 in inputs. This provides one of the foundations to make XCS evolve an optimal population of classifiers, where each classifier has the possibility to cover a set of perceptions. However, when performing XCS to solve the multi-step problems, i.e., maze control problems, the classifiers only allow the agent to perceive its surrounding environments without the direction information, which are contrary to our human perception. This paper develops an extension of XCS by introducing cyclic conditions to represent the classifiers. The proposed system, named generalised XCS classifier system (GXCS), is dedicated to modify the forms of the classifiers from {$\backslash$}emph\{chains\} to {$\backslash$}emph\{cycles\}, which allows them to match with more adjacent environments perceived by the agent from different directions. Accordingly, a more compact population of classifiers can be evolved to perform the generalisation feature of GXCS. As a first step of this research, GXCS has been tested on the benchmark maze control problems in which the agent can perceive its 8 surrounding cells. It is confirmed that GXCS can evolve the classifiers with cyclic conditions to successfully solve the problems as XCS, but with much smaller population size. }} @InProceedings{Lee:2014:CECb, title = {Applying {LCS} to Affective Images Classification in Spatial-Frequency Domain}, author = {Po-Ming Lee and Tzu-Chien Hsiao}, pages = {1690--1697}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Learning classifier systems, Classification, clustering and data analysis, Biometrics, bioinformatics and biomedical applications}, abstract = { Affective image classification is a task aims on classifying images based on their affective characteristics of inducing human emotions. This study accomplishes the task by using Linear Classifier System (LCS) and spatial frequency features. The model built by using LCS achieves Area Under Curve (AUC) = 0.91 and accuracy rate over 86\%. For comparison purposes, the result of the LCS is compared with other traditional machine learning classifiers (e.g., RBF Network) that are normally used in classification tasks. The study also presents user-independent results that indicate that the horizontal visual stimulations contribute more to emotion elicitation, than vertical visual stimulation. }} @InProceedings{Nguyen:2014:CECa, title = {A Novel Genetic Algorithm Approach for Simultaneous Feature and Classifier Selection in Multi Classifier System}, author = {Tien Thanh Nguyen and Alan Wee-Chung Liew and Minh Toan Tran and Xuan Cuong Pham and Mai Phuong Nguyen}, pages = {1698--1705}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Genetic algorithms, Learning classifier systems}, abstract = { In this paper we introduce a novel approach for classifier and feature selection in a multi-classifier system using Genetic Algorithm (GA). Specifically, we propose a 2-part structure for each chromosome in which the first part is encoding for classifier and the second part is encoding for feature. Our structure is simple in the implementation of the crossover as well as the mutation stage of GA. We also study 8 different fitness functions for our GA based algorithm to explore the optimal fitness functions for our model. Experiments are conducted on both 14 UCI Machine Learning Repository and CLEF2009 medical image database to demonstrate the benefit of our model on reducing classification error rate. }} @InProceedings{Glette:2014:CEC, title = {Lookup Table Partial Reconfiguration for an Evolvable Hardware Classifier System}, author = {Kyrre Glette and Paul Kaufmann}, pages = {1706--1713}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {EHW, Hardware Aspects of Bio-Inspired Architectures and Systems (HABIAS)}, abstract = { The evolvable hardware (EHW) paradigm relies on continuous run-time reconfiguration of hardware. When applied on modern FPGAs, the technically challenging reconfiguration process becomes an issue and can be approached at multiple levels. In related work, virtual reconfigurable circuits (VRC), partial reconfiguration, and lookup table (LUT) reconfiguration approaches have been investigated. In this paper, we show how fine-grained partial reconfiguration of 6-input LUTs of modern Xilinx FPGAs can lead to significantly more efficient resource use in an EHW application. Neither manual placement nor any proprietary bitstream manipulation is required in the simplest form of the employed method. We specify the goal architecture in VHDL and read out the locations of the automatically placed LUTs for use in an on line reconfiguration setting. This allows for an easy and flexible architecture specification, as well as possible implementation improvements over a hand-placed design. For demonstration, we rely on a hardware signal classifier application. Our results show that the proposed approach can fit a classification circuit 4 times larger than an equivalent VRC-based approach, and 6 times larger than a shift register-based approach, in a Xilinx Virtex-5 device. To verify the reconfiguration process, a MicroBlaze-based embedded system is implemented, and reconfiguration is carried out via the Xilinx Internal Configuration Access Port (ICAP) and driver software. }} % Session: ThE1-1 Ant Colony Optimisation @InProceedings{Pat:2014:CEC, title = {Ant Colony Optimization and Hypergraph Covering Problems}, author = {Ankit Pat}, pages = {1714--1720}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation, Discrete and combinatorial optimisation, Convergence, scalability and complexity analysis}, abstract = { Ant Colony Optimisation (ACO) is a very popular metaheuristic for solving computationally hard combinatorial optimisation problems. Runtime analysis of ACO with respect to various pseudo-Boolean functions and different graph based combinatorial optimisation problems has been taken up in recent years. In this paper, we investigate the runtime behaviour of an MMAS*(Max-Min Ant System) ACO algorithm on some well known hypergraph covering problems that are NP-Hard. In particular, we have addressed the Minimum Edge Cover problem, the Minimum Vertex Cover problem and the Maximum Weak-Independent Set problem. The influence of pheromone values and heuristic information on the running time is analysed. The results indicate that the heuristic information has greater impact towards improving the expected optimisation time as compared to pheromone values. For certain instances of hypergraphs, we show that the MMAS* algorithm gives a constant order expected optimisation time when the dominance of heuristic information is suitably increased. }} @InProceedings{He:2014:CECb, title = {Confidence-Based Ant Random Walks}, author = {Ping He and Ling Lu and Xiaohua Xu and Kanwen Li and Heng Qian and Wei Zhang}, pages = {1721--1728}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation, Classification, clustering and data analysis}, abstract = { To facilitate the computer-aided medical applications, this paper tries to build better intelligent diagnosis systems with the help of swarm intelligence method. As to the clinical data, a built-in graph structure is constructed with training samples being mapped as labelled vertices and test samples being unlabelled vertices. On the basis of the iterative label propagation algorithm, this paper first introduces a confidence-based random walk learning model, where unlabelled vertices that consistently show high probability (above the confidence threshold) in belonging to one class is treated as labelled vertices in the next iteration. Later motivated by the swarm intelligence, this model is further improved by treating the labelled vertices as real ants in nature and the predefined classes as different ant colonies. A novel labelled ant random walk algorithm is introduced by incorporating the history information of random walk in the form of aggregation pheromone. The proposed algorithms are evaluated with a synthetic data as well as some real-life clinical cases in terms of diagnostic accuracy. Experimental results show the potentiality of the proposed algorithms. }} @InProceedings{Kaszkurewicz:2014:CEC, title = {The Coupled {EigenAnt} Algorithm for Shortest Path Problems}, author = {Eugenius Kaszkurewicz and Amit Bhaya and Jayadeva Jayadeva and Joao Marcos Meirelles da Silva}, pages = {1729--1735}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Ant colony optimisation}, abstract = { This paper introduces an ACO model and associated algorithm, called Coupled Eigen Ant, for the problem of finding the shortest of \$N\$ paths between a source and a destination node. It is based on the recently introduced EigenAnt algorithm, the novelty being that it allows probabilistic path choice on both the forward and return journeys, as well as the fact that it introduces decay of pheromone deposition following a geometric progression. Equilibrium points of the model are calculated and the local stability of the two path synchronous version analysed. Simulations illustrate the main features of the algorithm. }} @InProceedings{Dawson:2014:CEC, title = {Accelerating Ant Colony Optimization-Based Edge Detection on the {GPU} Using {CUDA}}, author = {Laurence Dawson and Iain Stewart}, pages = {1736--1743}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Parallel and distributed algorithms, Ant colony optimisation}, abstract = { Ant Colony Optimisation (ACO) is a nature-inspired metaheuristic that can be applied to a wide range of optimisation problems. In this paper we present the first parallel implementation of an ACO-based (image processing) edge detection algorithm on the Graphics Processing Unit (GPU) using NVIDIA CUDA. We extend recent work so that we are able to implement a novel data-parallel approach that maps individual ants to thread warps. By exploiting the massively parallel nature of the GPU, we are able to execute significantly more ants per ACO-iteration allowing us to reduce the total number of iterations required to create an edge map. We hope that reducing the execution time of an ACO-based implementation of edge detection will increase its viability in image processing and computer vision. }} @InProceedings{Wu:2014:CECc, title = {Absorption in Model-Based Search Algorithms for Combinatorial Optimization}, author = {Zijun Wu and Michael Kolonko}, pages = {1744--1751}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Convergence, scalability and complexity analysis, Ant colony optimisation, Evolutionary programming}, abstract = { Model-based search is an abstract framework that unifies the main features of a large class of heuristic procedures for combinatorial optimisation, it includes ant algorithms, cross entropy and estimation of distribution algorithms. Properties shown for the model-based search therefore apply to all these algorithms. A crucial parameter for the long term behaviour of model-based search is the learning rate that controls the update of the model when new information from samples is available. Often this rate is kept constant over time. We show that in this case after finitely many iterations, all model-based search algorithms will be absorbed into a state where all samples consist of a single solution only. Moreover, it cannot be guaranteed that this solution is optimal, at least not when the optimal solution is unique. }} @InProceedings{Mavrovouniotis:2014:CECa, title = {Elitism-Based Immigrants for Ant Colony Optimization in Dynamic Environments: Adapting the Replacement Rate}, author = {Michalis Mavrovouniotis and Shengxiang Yang}, pages = {1752--1759}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments, Ant colony optimisation}, abstract = { The integration of immigrants schemes with ant colony optimisation (ACO) algorithms showed promising results on different dynamic optimisation problems (DOPs). The principle of integrating immigrants schemes within ACO is to introduce newly generated ants that will replace other ants in the current population. One of the most advanced immigrants schemes is the elitism-based immigrants scheme, where the best ant from the previous environment is used as the base to generate immigrants. So far, the replacement rate used for elitism-based immigrants in ACO remained fixed during the execution of the algorithm. In this paper the impact of the replacement rate on the performance of ACO algorithms with elitism-based immigrants is examined. In addition, an adaptive replacement rate is proposed and compared with fixed and optimised replacement rates based on a series of DOPs. The experiments show that the adaptive scheme provides an automatic way to set a good value, although not the optimal one, for the replacement rate within ACO with elitism-based immigrants for DOPs }} % Session: ThE1-2 Opposition-Based Learning and Differential Evolution @InProceedings{Mallipeddi:2014:CEC, title = {Gaussian Adaptation Based Parameter Adaptation for Differential Evolution}, author = {Rammohan Mallipeddi and Guohua Wu and Minho Lee and Suganthan Ponnuthurai Nagaratnam}, pages = {1760--1767}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Single Objective Numerical Optimisation}, abstract = { Differential Evolution (DE), a global optimisation algorithm based on the concepts of Darwinian evolution, is popular for its simplicity and effectiveness in solving numerous real-world optimisation problems in real valued spaces. The effectiveness of DE is due to the differential mutation operator that allows DE to automatically adjust between the exploration/exploitation in its search moves. However, the performance of DE is dependent on the setting of control parameters such as the mutation factor and the crossover probability. Therefore, to obtain optimal performance preliminary tuning of the numerical parameters, which is quite timing consuming, is needed. Recently, different parameter adaptation techniques, which can automatically update the control parameters to appropriate values to suit the characteristics of optimisation problems, have been proposed. However, most of the adaptation techniques try to adapt each of the parameter individually but do not take into account interaction between the parameters that are being adapted. In this paper, we introduce a DE self-adaptive scheme that takes into account the parameters dependencies by means of a multivariate probabilistic technique based on Gaussian Adaptation working on the parameter space. The performance of the DE algorithm with the proposed parameter adaptation scheme is evaluated on the benchmark problems designed for CEC 2014. }} @InProceedings{Salehinejad:2014:CEC, title = {Toward Using Type-{II} Opposition in Optimization}, author = {Hojjat Salehinejad and Shahryar Rahnamayan and Hamid R. Tizhoosh}, pages = {1768--1775}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Recent Advances on Opposition-Based Learning \& Applications, Differential evolution}, abstract = { The concept of opposition-based learning (OBL) can be categorised into Type-I and Type-II OBL methodologies. The Type-I OBL is based on the opposite points in the variable space while the Type-II OBL considers the opposite of function value on the landscape. In the past few years, many research works have been conducted on development of Type-I OBL-based approaches with application in science and engineering, such as opposition-based differential evolution (ODE). However, compared to Type-I OBL, which cannot address a real sense of opposition in term of objective value, the Type-II OBL is capable to discover more meaningful knowledge about problem's landscape. Due to natural difficulty of proposing a Type-II-based approach, very limited research has been reported in that direction. In this paper, for the first time, the concept of Type-II OBL has been investigated in detail in optimisation; also it is applied on the DE algorithm as a case study. The proposed algorithm is called opposition based differential evolution Type-II (ODE-II) algorithm; it is validated on the testbed proposed for the IEEE Congress on Evolutionary Computation 2013 (IEEE CEC-2013) contest with 28 benchmark functions. Simulation results on the benchmark functions demonstrate the effectiveness of the proposed method as the first step for further developments in Type-II OBL-based schemes. }} @InProceedings{Liu:2014:CECj, title = {Improved Differential Evolution with Adaptive Opposition Strategy}, author = {Huichao Liu and Zhijian Wu and Hui Wang and Shahryar Rahnamayan and Changshou Deng}, pages = {1776--1783}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Recent Advances on Opposition-Based Learning \& Applications, Differential evolution}, abstract = { Generalised opposition-based differential evolution (GODE) is an effective algorithm for global optimisation over continuous search space. However, the performance of GODE highly depends on its control parameters. To improve the performance of GODE, this paper proposes an enhanced GODE algorithm called AGODE, which employs an adaptive generalised opposition-based learning (GOBL) mechanism to automatically adjust the probability of opposition during the evolution. Experimental study is conducted on a set of 19 well known benchmark functions. Computational results show that the proposed approach AGODE outperforms some state-of-the-art DE variants on the majority of test problems. }} @InProceedings{Angelo:2014:CEC, title = {Differential Evolution Assisted by a Surrogate Model for Bilevel Programming Problems}, author = {Jaqueline Angelo and Eduardo Krempser and Helio Barbosa}, pages = {1784--1791}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Bilevel Optimisation, Differential evolution, Meta-modelling and surrogate models}, abstract = { Bilevel programming is used to model decentralised problems involving two levels of decision makers that are hierarchically related. Those problems, which arise in many practical applications, are recognised to be challenging. This paper reports a Differential Evolution (DE) method assisted by a surrogate model to solve bilevel programming problems(BLPs). The method proposed is an extension of a previous one, BlDE, developed by the authors, where two DE methods are used to generate and evolve the upper and the lower level variables. Here, the use of a similarity-based surrogate model and a different stopping criteria are proposed in order to reduce the number of function evaluations on both levels of the bilevel optimisation. The numerical results show a significant reduction in the number of function evaluations in the lower level of the problem. }} @InProceedings{Minisci:2014:CEC, title = {Adaptive Inflationary Differential Evolution}, author = {Edmondo Minisci and Massimiliano Vasile}, pages = {1792--1799}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Self-adaptation in evolutionary computation, Differential evolution, Real-world applications}, abstract = { In this paper, an adaptive version of Inflationary Differential Evolution is presented and tested on a set of real case problems taken from the CEC2011 competition on real-world applications. Inflationary Differential Evolution extends standard Differential Evolution with both local and global restart procedures. The proposed adaptive algorithm uses a probabilistic kernel based approach to automatically adapt the values of both the crossover and step parameters. In addition the paper presents a sensitivity analysis on the values of the parameters controlling the local restart mechanism and their impact on the solution of one of the hardest problems in the CEC2011 test set. }} @InProceedings{Rahnamayan:2014:CEC, title = {Computing Opposition by Involving Entire Population}, author = {Shahryar Rahnamayan and Jude Jesuthasan and Faird Bourennani and Hojjat Salehinejad and Greg F. Naterer}, pages = {1800--1807}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Recent Advances on Opposition-Based Learning \& Applications, Differential evolution}, abstract = { The capabilities of evolutionary algorithms (EAs) in solving nonlinear and non-convex optimisation problems are significant. Among the many types of methods, differential evolution (DE) is an effective population-based stochastic algorithm, which has emerged as very competitive. Since its inception in 1995, many variants of DE to improve the performance of its predecessor have been introduced. In this context, opposition-based differential evolution (ODE) established a novel concept in which, each individual must compete with its opposite in terms of the fitness value in order to make an entry in the next generation. The generation of opposite points is based on the population's current extreme points (i.e., maximum and minimum) in the search space; these extreme points are not proper representatives for whole population, compared to centroid point which is inclusive regarding all individuals in the population. This paper develops a new scheme that uses the centroid point of a population to calculate opposite individuals. Therefore, the classical scheme of an opposite point is modified accordingly. Incorporating this new scheme into ODE leads to an enhanced ODE that is identified as centroid opposition-based differential evolution (CODE). The performance of the CODE algorithm is comprehensively evaluated on well-known complex benchmark functions and compared with the performance of conventional DE, ODE, and some other state-of-the-art algorithms (such as SaDE, ADE, SDE, and jDE) in terms of solution accuracy. The results for CODE are promising. }} % Session: ThE1-3 Genetic Programming @InProceedings{Li:2014:CECk, title = {Adaptive Genetic Network Programming}, author = {Xianneng Li and Wen He and Kotaro Hirasawa}, pages = {1808--1815}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Genetic Network Programming, Self-adaptation in evolutionary computation, Intelligent systems applications}, abstract = { Genetic Network Programming (GNP) is derived from Genetic Algorithm (GA) and Genetic Programming (GP), which applies evolution theory to evolve a population of directed graph to model complex systems. It has been shown that GNP can solve typical control problems, as well as many real-world problems. However, studying GNP is mainly focused on the specific aspect, while the fundamental characteristics that ensure the success of GNP are rarely investigated in the previous research. This paper reveals an important feature of GNP, reusability of nodes, to efficiently identify and formulate the building blocks of evolution. Accordingly, adaptive GNP is developed which self-adapts both crossover and mutation probabilities of each search variable to circumstances. The adaptation allows the automatic adjustment of evolution bias toward the frequently reused nodes in high-quality individuals. The adaptive GNP is compared with traditional GNP in a benchmark control testbed to evaluate its superiority. }} @InProceedings{Weise:2014:CEC, title = {Evolving Exact Integer Algorithms with Genetic Programming}, author = {Thomas Weise and Mingxu Wan and Ke Tang and Xin Yao}, pages = {1816--1823}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Representation and operators}, abstract = { The synthesis of exact integer algorithms is a hard task for Genetic Programming (GP), as it exhibits epistasis and deceptiveness. Most existing studies in this domain only target few and simple problems or test a small set of different representations. In this paper, we present the (to the best of our knowledge) largest study on this domain to date. We first propose a novel benchmark suite of 20 non-trivial problems with a variety of different features. We then test two approaches to reduce the impact of the negative features: (a) a new nested form of Transactional Memory (TM) to reduce epistatic effects by allowing instructions in the program code to be permutated with less impact on the program behaviour and (b) our recently published Frequency Fitness Assignment method (FFA) to reduce the chance of premature convergence on deceptive problems. In a full-factorial experiment with six different loop instructions, TM, and FFA, we find that GP is able to solve all benchmark problems, although not all of them with a high success rate. Several interesting algorithms are discovered. FFA has a tremendous positive impact while TM turns out not to be useful. }} @InProceedings{Nguyen:2014:CECb, title = {A Sequential Genetic Programming Method to Learn Forward Construction Heuristics for Order Acceptance and Scheduling}, author = {Su Nguyen and Mengjie Zhang and Mark Johnston}, pages = {1824--1831}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Evolutionary Computation for Planning and Scheduling, Heuristic Methods for Multi-Component Optimisation Problems}, abstract = { Order acceptance and scheduling (OAS) is a hard optimisation problem in which both acceptance decisions and scheduling decisions must be considered simultaneously. Designing effective solution methods or heuristics for OAS is not a trivial task, especially to deal with different problem configurations and sizes. This paper proposes a new heuristic framework called forward construction heuristic (FCH) for OAS and develops a new sequential genetic programming (SGPOAS) method for automatic design of FCHs. The key idea of the new GP method is to learn priority rules directly from optimal scheduling decisions at different decision moments and evolve a set of rules for FCHs instead of a single rule as shown in previous studies. The results show that evolved FCHs are significantly better than evolved single priority rules. The evolved FCHs are also competitive with the existing meta-heuristics in the literature and very effective for large problem instances. }} @InProceedings{Xie:2014:CEC, title = {Anomaly Detection in Crowded Scenes Using Genetic Programming}, author = {Cheng Xie and Lin Shang}, pages = {1832--1839}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Real-world applications}, abstract = { Genetic programming(GP) has become an increasingly hot issue in evolutionary computation due to its extensive application. Anomaly detection in crowded scenes is also a hot research topic in computer vision. However, there are few contributions on using genetic programming to detect abnormalities in crowded scenes. In this paper, we focus on anomaly detection in crowded scenes with genetic programming. We propose a new method called Multi-Frame LBP Difference (MFLD) based on Local Binary Patterns(LBP) to extract pixel-level features from videos without additional complicated preprocessing operations such as optical flow and background subtraction. Genetic programming is employed to generate an anomaly detector with the extracted data. When a new video is coming, the detector can classify every frame and localise the abnormality to a single pixel level in real time. We validate our approach on a public dataset and compare our method with other traditional algorithms for video anomaly detection. Experimental results indicate that our method with genetic programming performs better in detecting abnormalities in crowded scenes. }} @InProceedings{Yu:2014:CECe, title = {A Genetic Programming Approach to Distributed {QoS}-Aware Web Service Composition}, author = {Yang Yu and Hui Ma and Mengjie Zhang}, pages = {1840--1846}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Parallel and Distributed Evolutionary Computation in the Cloud Era}, abstract = { Web service composition has emerged as a promising technique for building complex web applications, thus supporting business-to-business and enterprise application integration. Nowadays there are increasing numbers of web services are distributed across the Internet. For a given service request there are many ways of service composition that can meet the service functional requirements (inputs and outputs) but have different qualities of Services (QoS), like response time or execution cost. QoS-aware web service composition seeks to find a service composition with optimised QoS properties. Genetic Programming is an efficient tool for tacking such optimisation problems efficiently. This paper proposes a novel GP-based approach for distributed web service composition where multiple QoS constraints are considered simultaneously. A series of experiments have been conducted to evaluate the proposed approach with test data. The results show that our approach is efficient and effective to find a near-optimal service composition solution in the context of distributed service environment. }} @InProceedings{Kren:2014:CEC, title = {Generating Lambda Term Individuals in Typed Genetic Programming Using Forgetful {A*}}, author = {Tomas Kren and Roman Neruda}, pages = {1847--1854}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Representation and operators, lambda calculus, initialisation}, abstract = { Tree based genetic programming (GP) traditionally uses simple S-expressions to represent programs, however more expressive representations, such as lambda calculus, can exhibit better results while being better suited for typed GP. In this paper we present population initialisation methods within a framework of GP over simply typed lambda calculus that can be also used in the standard GP approach. Initialisations can be parametrised by different search strategies, leading to wide spectrum of methods corresponding to standard ramped half-and-half initialisation on one hand, or exhaustive systematic search on the other. A novel geometric strategy is proposed that balances those two approaches. Experiments on well known benchmark problems show that the geometric strategy outperforms the standard generating method in success rate, best fitness value, time consumption and average individual size. }} % Session: ThE1-4 Heuristics, Metaheuristics and Hyper-heuristics I @InProceedings{Perdigao-Cota:2014:CEC, title = {{AIRP}: A Heuristic Algorithm for Solving the Unrelated Parallel Machine Scheduling Problem}, author = {Luciano Perdigao Cota and Matheus Nohra Haddad and Marcone Jamilson Freitas Souza and Vitor Nazario Coelho}, pages = {1855--1862}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Discrete and combinatorial optimisation}, abstract = { This paper deals with the Unrelated Parallel Machine Scheduling Problem with Setup Times (UPMSPST). The objective is to minimise the makespan. In order to solve it, we propose a heuristic algorithm, based on Iterated Local Search (ILS), Variable Neighbourhood Descent (VND) and Path Relinking (PR). In this algorithm, named AIRP, an initial solution is constructed using the Adaptive Shortest Processing Time method. This solution is refined by the ILS, having an adaptation of the VND as local search method. The PR method is applied as a strategy of intensification and diversification during the search. The algorithm was tested in instances of the literature involving up to 150 jobs and 20 machines. The computational experiments show that the proposed algorithm outperforms an algorithm from the literature, both in terms of quality and variability of the final solution. In addition, the algorithm established new best solutions for more than 80,5\% of the test problems in average. }} @InProceedings{Grobler:2014:CEC, title = {Heuristic Space Diversity Management in a Meta-Hyper-Heuristic Framework}, author = {Jacomine Grobler and Andries P. Engelbrecht and Graham Kendall and V.S.S. Yadavalli}, pages = {1863--1869}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics}, abstract = { This paper introduces the concept of heuristic space diversity and investigates various strategies for the management of heuristic space diversity within the context of a meta-hyper-heuristic algorithm. Evaluation on a diverse set of floating-point benchmark problems show that heuristic space diversity has a significant impact on hyper-heuristic performance. The increasing heuristic space diversity strategies performed the best out of all strategies tested. Good performance was also demonstrated with respect to another popular multi-method algorithm and the best performing constituent algorithm. }} @InProceedings{Sinha:2014:CEC, title = {An Improved Bilevel Evolutionary Algorithm Based on Quadratic Approximations}, author = {Ankur Sinha and Pekka Malo and Kalyanmoy Deb}, pages = {1870--1877}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Bilevel Optimisation}, abstract = { In this paper, we provide an improved evolutionary algorithm for bilevel optimisation. It is an extension of a recently proposed Bilevel Evolutionary Algorithm based on Quadratic Approximations (BLEAQ). Bilevel optimisation problems are known to be difficult and computationally demanding. The recently proposed BLEAQ approach has been able to bring down the computational expense significantly as compared to the contemporary approaches. The strategy proposed in this paper further improves the algorithm by incorporating archiving and local search. Archiving is used to store the feasible members produced during the course of the algorithm that provide a larger pool of members for better quadratic approximations of optimal lower level solutions. Frequent local searches at upper level supported by the quadratic approximations help in faster convergence of the algorithm. The improved results have been demonstrated on two different sets of test problems, and comparison results against the contemporary approaches are also provided. }} @InProceedings{Ke:2014:CEC, title = {A Cooperative Approach between Metaheuristic and Branch-and-Price for the Team Orienteering Problem with Time Windows}, author = {Liangjun Ke}, pages = {1878--1882}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Discrete and combinatorial optimisation}, abstract = { The team orienteering problem with time windows (TOPTW) is a well studied routing problem. In this paper, a cooperative algorithm is proposed. It collaborates metaheuristic and branch-and-price. A restricted master problem and subproblem are defined. It uses a heuristic to obtain an integral solution for the restricted master problem and a metaheuristic to generate new columns for the subproblem. Experimental study shows that this algorithm can find new better solutions for several instances in short time, which supports the effectiveness of the cooperative mechanism between metaheuristic and branch-and-price. }} @InProceedings{Zheng:2014:CECc, title = {Hyper-Heuristics with Penalty Parameter Adaptation for Constrained Optimization}, author = {Yu-Jun Zheng and Bei Zhang and Zhen Cheng}, pages = {1883--1889}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Constraint handling}, abstract = { Penalty functions are widely used in constrained optimisation, but determining optimal penalty parameters or weights turns out to be a difficult optimisation problem itself. The paper proposes a hyper-heuristic approach, which searches the optimal penalty weight setting for low-level heuristics, taking the performance of those heuristics with specialised penalty weight settings as feedback to adjust the high-level search. The proposed approach can either be used for merely improving low-level heuristics, or be combined into a common hyper-heuristic framework for constrained optimisation. Experiments on a set of well-known benchmark problems show that the hyper-heuristic approach with penalty parameter adaptation is effective in both aspects. }} @InProceedings{Segredo:2014:CEC, title = {Control of Numeric and Symbolic Parameters with a Hybrid Scheme Based on Fuzzy Logic and Hyper-heuristics}, author = {Eduardo Segredo and Carlos Segura and Coromoto Leon}, pages = {1890--1897}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Multi-objective evolutionary algorithms, Large-scale problems}, abstract = { One of the main disadvantages of Evolutionary Algorithms (EAs) is that they converge towards local optima for some problems. In recent years, diversity based multi-objective EAs have emerged as a promising technique to prevent from local optima stagnation when optimising single-objective problems. An additional drawback of EAs is the large dependency between the quality of the results provided and the setting of their parameters. By the use of parameter control methods, parameter values can be adapted during the run of an EA. The aim of control approaches is not only to improve the robustness of the controlled algorithm, but also to boost its efficiency. In this paper we apply a novel hybrid parameter control scheme based on Fuzzy Logic and Hyper-heuristics to simultaneously adapt several numeric and symbolic parameters of a diversity based multi-objective EA. An extensive experimental evaluation is carried out, which includes a comparison between the hybrid control proposal and a wide range of configurations of the diversity-based multi-objective EA with fixed parameters. Results demonstrate that our control proposal is able to find similar or even better solutions than those obtained by the best configuration of the diversity-based scheme with fixed parameters in a significant number of benchmark problems, demonstrating the advantages of parameter control over parameter tuning for these test cases. }} % Special Track: Industrial Session: ThE1-5 Computational Intelligence on Predictive Maintenance and Optimisation @InProceedings{Sayed:2014:CEC, title = {A Decomposition-Based Algorithm for Dynamic Economic Dispatch Problems}, author = {Eman Sayed and Daryl Essam and Ruhul Sarker and Saber Elsayed}, pages = {1898--1905}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems, Real-world applications, Computational Intelligence for Industrial Applications}, abstract = { Large scale constrained problems are complex problems due to their dimensionality, structure, in addition to their constraints. The performance of EAs decreases when the problem dimension increases. Decomposition-based EAs can overcome this drawback, but their performance would be affected if the interdependent variables were optimised in different subproblems. The use of EAs with variables interaction identification technique handles this issue by identifying better arrangements for decomposing a large problem into subproblems in a way that minimises the interdependencies between them. The only technique in the literature that has been developed to identify the variables interdependency in constrained problems is the Variable Interaction Identification for Constrained problems (VIIC). This technique is tested in this paper on a real-world problem at three large dimensions which are large scale constrained optimisation problems. The performance of the decomposition-based EA that uses VIIC is compared to Random Grouping approach for decomposition, for 5-Units, 10-Units, and 30-Units DED problems. }} @InProceedings{Ding:2014:CEC, title = {Minimizing Makespan for a No-Wait Flowshop Using Tabu Mechanism Improved Iterated Greedy Algorithm}, author = {Jianya Ding and Shiji Song and Rui Zhang and Cheng Wu}, pages = {1906--1911}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence for Industrial Applications, Discrete and combinatorial optimisation}, abstract = { This paper proposes a tabu mechanism improved iterated greedy (TMIIG) algorithm to solve the no-wait flowshop scheduling problem with makespan criterion. The motivation of seeking for further improvement in the iterated greedy (IG) algorithm framework is based on the observation that the construction phase of the original IG algorithm may lead to repeated search when applying the insertion neighbourhood search. To overcome the drawback, we modified the IG algorithm by a tabu-based reconstruction strategy to enhance its exploitation ability. A powerful neighbourhood search method which involves insert, swap, and double-insert moves is then applied to obtain better solutions from the reconstructed solution in the previous step. Numerical computations verified the advantages of the new reconstruction scheme. In addition, comparisons with other high-performing algorithms demonstrated the effectiveness and robustness of the proposed algorithm. }} @InProceedings{Ruello:2014:CEC, title = {Black-Hole {PSO} and {SNO} for Electromagnetic Optimization}, author = {Matteo Ruello and Francesco Grimaccia and Marco Mussetta and Riccardo E. Zich}, pages = {1912--1916}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Real-world applications, Computational Intelligence for Industrial Applications}, abstract = { In the past years Particle Swarm Optimisation (PSO) has gained increasing attention for engineering and real-world applications. Among these, the design of antennas and electromagnetic devices is a well established field of application. More recently, Social Network Optimisation (SNO) has been introduced, inspired by the recent explosion of social networks and their capability to drive people's decision making process in everyday life. ``Black-hole'' is a novel operator, which is here considered for both PSO and SNO. It is based on the concept of repulsion among agents when they get stuck in local optima. The design of a planar array antenna is here addressed in order to assess its performances on a benchmark EM optimisation problem. Reported results show its effectiveness in dealing with antenna optimisation. }} @InProceedings{Qian:2014:CEC, title = {An Improved Ant Colony Algorithm for Winner Determination in Multi-Attribute Combinatorial Reverse Auction}, author = {Xiaohu Qian and Min Huang and Taiguang Gao and Xingwei Wang}, pages = {1917--1921}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence for Industrial Applications, Ant colony optimisation, Discrete and combinatorial optimisation}, abstract = { This paper considers the problem of one buyer procuring multi-items from multiple potential suppliers in the electronic reverse auction, where each supplier can bid on combinations of items. From the perspective of the buyer, by considering multi-attributes of each item, a winner determination problem (WDP) of multi-items single-unit combinatorial reverse auctions was described and a bi-objective programming model was established. According to the characteristics of the model, an equivalent single-objective programming model was obtained. As the problem is NP-hard, an improved ant colony (IAC) algorithm considering the dynamic transition strategy and the Max-Min pheromone strategy is proposed for the problem. Experimental results show the effectiveness of the improved algorithm. }} @InProceedings{Pandiyan:2014:CEC, title = {Soft Computing Techniques Based Optimal Tuning of Virtual Feedback {PID} Controller for Chemical Tank Reactor}, author = {Manikandan Pandiyan}, pages = {1922--1928}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolution strategies, Ant colony optimisation, Particle swarm optimisation (PSO)}, abstract = { CSTR plays a vital role in almost all the chemical reactions and is a highly nonlinear system exhibiting stable as well as unstable steady states. The variables which characterise the quality of the final product in CSTR are often difficult to measure in real-time and cannot be directly measured using the feedback configuration [1]. So, a virtual feedback control is implemented to control the state variables using Extended Kalman Filter (EKF) in the feedback path. Since it is hard to determine the optimal or near optimal PID parameters using classical tuning techniques like Ziegler Nichols method, a highly skilled optimisation algorithm like Particle Swarm Optimisation (PSO) and Ant Colony Optimisation (ACO) are used. This work is based on the optimal tuning of virtual feedback PID control for a CSTR system using soft computing algorithm for minimum Integral Square Error (ISE) condition. }} % Plenary Poster Session: PE4 Poster Session IV @InProceedings{Harrison:2014:CEC, title = {Dynamic Multi-Objective Optimization Using Charged Vector Evaluated Particle Swarm Optimization}, author = {Kyle Harrison and Beatrice Ombuki-Berman and Andries Engelbrecht}, pages = {1929--1936}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Dynamic Multi-objective Optimisation}, abstract = { The vector evaluated particle swarm optimisation (VEPSO) algorithm is a multi-swarm variation of the traditional particle swarm optimisation (PSO) used to solve static multi-objective optimisation problems (MOOPs). Recently, the dynamic VEPSO (DVEPSO) algorithm was proposed as an extension to VEPSO enabling the algorithm to handle dynamic MOOPs (DMOOPs). While DVEPSO has been successful at handling DMOOPs, the change detection mechanism relied on observing changes in objective space. An alternative strategy is proposed by using charged PSO (CPSO) sub-swarms with decision space change detection to address the outdated memory issue observed in vanilla PSO. This dynamic PSO variant allows for (implicit) decision space tracking not seen in DVEPSO while implicitly handling the diversity issue seen in dynamic environments. The proposed charged VEPSO is compared to DVEPSO on a wide variety of dynamic environment types. Results indicated that, in general, the proposed charged VEPSO outperformed the existing DVEPSO. Further, charged VEPSO exhibited better front-tracking abilities, while DVEPSO was superior with regards to locating the Pareto front. }} @InProceedings{Mesa:2014:CEC, title = {A New Self-Adaptive {PSO} Based on the Identification of Planar Regions}, author = {Eddy Mesa and Juan David Velasquez and Patricia Jaramillo}, pages = {1937--1943}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Self-adaptation in evolutionary computation}, abstract = { In this paper, we propose a new approach for self-adaptive particle swarm optimisation, using the function's topology to adapt the parameters and modifying them when a planar region is identified in the objective function. Particle swarm optimisation is a metaheuristic developed to optimise nonlinear problems. This metaheuristic has four parameters to adapt the search for the different optimisation problems. However, finding an optimal set of parameters is not a trivial problem. Some strategies to adapt the parameters have been developed, but they are not robust enough to cover all kinds of problems. Function's topology is one of the most decisive factors in order to choose a right set of parameters; i.e. convex functions need more exploitation because this topology offers a clear direction to the minimum point. In the opposite way, a noise function can be trapped in a local minimum for the same level of exploitation. In order to validate and compare our methods, we use the benchmark functions from CEC 2005 to compare the different particle swarm optimisation versions. The results show that the proposed version is significant better than the original particle swarm optimisation and the standard particle swarm optimisation proposed in 2011. }} @InProceedings{Tsai:2014:CEC, title = {{PSO}-Based Evacuation Simulation Framework}, author = {Pei-Chuan Tsai and Chih-Ming Chen and Ying-ping Chen}, pages = {1944--1950}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Evolutionary games and multi-agent systems, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Evacuation simulation is a critical and important research issue for people to design safer building layouts or plan more effective evacuation routes. Many studies adopted methodologies in evolutionary computation into the evacuation simulation systems for finding better solutions. To simulate human behaviour or crowd motion is one key factor to the practicality of the system. Particle swarm optimisation algorithm (PSO), which is originated from the inspiration of bird flocking, is commonly applied to model human behaviour. Based on the PSO based human behaviour simulation, many studies have got good results on evacuation simulation. However, the configurations of describing the experiment environment in the literature are complicated and specialised for certain specific scenarios. Observing the fact, we propose a new PSO-based simulation framework in order to provide a simple and general way to configure various simulation scenarios. This work adopts our previously proposed PSO-based crowd movement controlling mechanism and introduces new mechanisms to make the simulation fitting into evacuation circumstance more real. In the proposed framework, all people, obstacles, exits, and even the evacuation guide indicators are modelled as the original component of the PSO algorithm. It is convenient to setup the simulation environment upon the framework. Therefore, taking the proposed work as a research tool will be advantageous when the issue of evacuation simulation is investigated. }} @InProceedings{Bouaziz:2014:CEC, title = {{PSO}-Based Update Memory for Improved Harmony Search Algorithm to the Evolution of {FBBFNT'} Parameters}, author = {Souhir Bouaziz and Adel M. Alimi and Ajith Abraham}, pages = {1951--1958}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Extended Genetic Programming, Memetic, multi-meme and hybrid algorithms}, abstract = { In this paper, a PSO-based update memory for Improved Harmony Search (PSOUM-IHS) algorithm is proposed to learn the parameters of Flexible Beta Basis Function Neural Tree (FBBFNT) model. These parameters are the Beta parameters of each flexible node and the connected weights of the network. Furthermore, the FBBFNT's structure is generated and optimised by the Extended Genetic Programming (EGP) algorithm. The combination of the PSOUM-IHS and EGP in the same algorithm is so used to evolve the FBBFNT model. The performance of the proposed evolving neural network is evaluated for nonlinear systems of prediction and identification and then compared with those of related models. }} @InProceedings{Jariyatantiwait:2014:CEC, title = {Fuzzy Multiobjective Differential Evolution Using Performance Metrics Feedback}, author = {Chatkaew Jariyatantiwait and Gary Yen}, pages = {1959--1966}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Multi-objective evolutionary algorithms}, abstract = { Differential evolution is regarded as one of the most efficient evolutionary algorithms to tackle multiobjective optimisation problems. The key to success of any multiobjective evolutionary algorithms (MOEAs) is maintaining a delicate balance between exploration and exploitation throughout the evolution process. In this paper, we propose a Fuzzy-based Multiobjective Differential Evolution (FMDE) that uses performance metrics, specifically, hypervolume, spacing, and maximum spread, to measure the state of the evolution process. We apply the inference rules to these metrics in order to dynamically adjust the associated control parameters of a chosen mutation strategy used in this algorithm. One parameter controls the degree of greedy or exploitation, while another regulates the degree of diversity or exploration of the reproduction phase. Therefore, we can appropriately adjust the degree of exploration and exploitation through performance feedback. The performance of FMDE is evaluated on well-known ZDT and DTLZ test suites in addition two representative functions in WFG. The results show that the proposed algorithm is competitive with respect to chosen state-of-the-art MOEAs. }} @InProceedings{Yuen:2014:CEC, title = {Multiobjective Evolutionary Algorithm Portfolio: Choosing Suitable Algorithm for Multiobjective Optimization Problem}, author = {Shiu Yin Yuen and Xin Zhang}, pages = {1967--1973}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Heuristics, metaheuristics and hyper-heuristics}, abstract = { The concept of algorithm portfolio has a long history. Recently this concept draws increasing attention from researchers, though most of the researches have concentrated on single objective optimisation problems. This paper is intended to solve multiobjective optimisation problems by proposing a multiple evolutionary algorithm portfolio. Differing from previous approaches, each component algorithm in our portfolio method has an independent population and the component algorithms do not communicate in any way with each other. Another difference is that our algorithm introduces no control parameters. This parameter-less characteristic is desirable as each additional parameter requires independent parameter tuning or control. A novel score calculation method,based on predicted performance, is used to assess the contributions of component algorithms during the optimisation process. Such information is used by an algorithm selector which decides, for each generation, which algorithm to use. Experimental results show that our portfolio method outperforms individual algorithms in the portfolio. Moreover, it outperforms the AMALGAM method. }} @InProceedings{Shang:2014:CEC, title = {A Novel Algorithm for Many-Objective Dimension Reductions: {Pareto-PCA-NSGA-II}}, author = {Ronghua Shang and Kun Zhang and Licheng Jiao}, pages = {1974--1981}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Multiobjective optimisation, Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { Many-objective problem has more than 3 objectives. Because of the extraordinary difficulty of acquiring their Pareto optimal solutions directly, traditional methods will be out of operation for such problems. In recent years, many researchers have turned their attention to the study of this area. They are interested in two areas: acquiring some part of Pareto front which is useful to the researchers (Preferred Solutions) and reducing redundant objectives. In this paper, we combine two dimension reduction methods: the method based on Pareto optimal solution analysis and the method based on correlation analysis, to form a novel algorithm for dimension reduction. Firstly, the Pareto optimal solutions are acquired through NSGA-II. Then the objectives who contribute little to the number of non-dominated solutions are removed. At last, the dimension of objectives is reduced further according to their contribution to the principal component in PCA analysis. In this way, we can acquire the right non-redundant objectives with low time complexity. Simulation results show that the proposed algorithm can effectively reduce redundant objectives and keep the non-redundant objectives with low time. }} @InProceedings{Souza:2014:CEC, title = {An Experimental Analysis of Evolutionary Algorithms for the Three-Objective Oil Derivatives Distribution Problem}, author = {Thatiana Souza and Elizabeth Goldbarg and Marco Goldbarg}, pages = {1982--1989}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial optimisation, Multi-objective evolutionary algorithms}, abstract = { Scheduling oil derivatives distribution by multi-product pipelines is an important problem faced by the petroleum industry. Some researchers deal with it as a discrete problem where batches of products flow in a network. Minimising delivery time is a usual objective handled by engineers when dealing with this problem. Nevertheless, other important costs may also be considered such as losses due to interfaces between fluids and electrical energy. Losses due to interfaces occur when different products sent consecutively contaminate each other. The price paid for electrical energy varies during the day, so it is important also to try to minimise this cost. In this paper, these three objectives, i.e. delivery time, interface losses and electricity cost, are minimised simultaneously. Two hybridisation of transgenetic algorithms with well-known multi-objective evolutionary algorithms are proposed. One is derived from the NSGA-II framework, named NSTA, and the other is derived from the MOEA/D framework, named MOTA/D. To analyse the performance of the proposed algorithms, they are compared with their classical counterparts and applied to thirty random instances. It is also the first time MOEA/D is applied to the investigated problem. Statistical tests indicate that the MOTA/D generated better approximation sets than the other algorithms. Therefore, the MOTA/D encourages further researches in the hybridisation of transgenetic algorithms and evolutionary multi-objective frameworks, specifically those based on decomposition. }} @InProceedings{Leung:2014:CEC, title = {A New Strategy for Finding Good Local Guides in {MOPSO}}, author = {Man Fai Leung and Sin Chun Ng and Chi Chung Cheung and Andrew K Lui}, pages = {1990--1997}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Particle swarm optimisation (PSO)}, abstract = { This paper presents a new algorithm that extends Particle Swarm Optimisation (PSO) to deal with multi-objective problems. It makes two main contributions. The first is that the square root distance (SRD) computation among particles and leaders is proposed to be the criterion of the local best selection. This new criterion can make all swarms explore the whole Pareto-front more uniformly. The second contribution is the procedure to update the archive members. When the external archive is full and a new member is to be added, an existing archive member with the smallest SRD value among its neighbours will be deleted. With this arrangement, the non-dominated solutions can be well distributed. Through the performance investigation, our proposed algorithm performed better than two well known multi-objective PSO algorithms, MOPSO-sigma and MOPSO-CD, in terms of different standard measures. }} @InProceedings{Yu:2014:CECf, title = {An Inter-Molecular Adaptive Collision Scheme for Chemical Reaction Optimization}, author = {James J.Q. Yu and Victor O.K. Li and Albert Y.S. Lam}, pages = {1998--2004}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Representation and operators}, abstract = { Optimisation techniques are frequently applied in science and engineering research and development. Evolutionary algorithms, as a kind of general-purpose metaheuristic, have been shown to be very effective in solving a wide range of optimisation problems. A recently proposed chemical-reaction inspired metaheuristic, Chemical Reaction Optimisation (CRO), has been applied to solve many global Optimisation problems. However, the functionality of the inter-molecular ineffective collision operator in the canonical CRO design overlaps that of the on-wall ineffective collision operator, which can potential impair the overall performance. In this paper we propose a new inter-molecular ineffective collision operator for CRO for global Optimisation. To fully our newly proposed operator, we also design a scheme to adapt the algorithm to Optimisation problems with different search space characteristics. We analyse the performance of our proposed algorithm with a number of widely used benchmark functions. The simulation results indicate that the new algorithm has superior performance over the canonical CRO. }} @InProceedings{Poole:2014:CECa, title = {Analysis of Constraint Handling Methods for the Gravitational Search Algorithm}, author = {Daniel Poole and Christian Allen and Thomas Rendall}, pages = {2005--2012}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling}, abstract = { The gravitational search algorithm (GSA) is a recent addition to the family of global Optimisation algorithms based on phenomena found in nature, specifically the gravitational attractive force between two bodies of mass. However, like almost all global search algorithms of this type, GSA has no direct method of handling a constrained Optimisation problem. There has been much attention to constraint handling using other agent based systems, though the mechanics of GSA make the application of many of these difficult. This paper has therefore analysed constraint handling methods for use with GSA and compared the performance of simple to implement methods (penalties and feasible directions) with a novel separation-sub-swarm (3S) approach, and found that feasible direction methods ideally need at least one initially feasible particle, and that the novel 3S approach is highly effective for solving constrained Optimisation problems using GSA outperforming the other approaches tested. }} @InProceedings{Cai:2014:CECa, title = {Distributed Wireless Sensor Scheduling for Multi-Target Tracking Based on Matrix-Coded Parallel Genetic Algorithm}, author = {Zixing Cai and Sha Wen and Lijue Liu}, pages = {2013--2018}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Parallel and distributed algorithms, Genetic algorithms, Multiobjective Optimisation}, abstract = { The aim of designing a sensor scheduling scheme for target tracking in wireless sensor network is to improve the tracking accuracy, balance the network energy and prolong the network lifespan. It is viewed as a multi-objective Optimisation problem. A modified matrix-coded parallel genetic algorithm (MPGA) is proposed in which multiple subpopulations evolve synchronously and satisfy the specific constraint arisen from the scenario of multi-target tracking that a sensor can only track just one target. Simulation results show that MPGA , compared with traditional genetic algorithm, converges to the better result with higher speed when applied in multi-target tracking in wireless sensor network. And our proposed distributed sensor scheduling scheme based on MPGA outperforms than existed schemes. }} @InProceedings{Ding:2014:CECa, title = {Effect of Pseudo Gradient on Differential Evolutionary for Global Numerical Optimization}, author = {Jinliang Ding and Lipeng Chen and Qingguang Xie and Tianyou Chai and Xiuping Zheng}, pages = {2019--2026}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Self-adaptation in evolutionary computation}, abstract = { In this paper, a novel pseudo gradient based DE approach is proposed, which takes advantage of both the differential evolutionary (DE) and the gradient-based algorithm. The gradient information, which is called pseudo gradient, is generated through randomly selected two vectors and their fitness function values. This work is to investigate the effect of proposed pseudo gradient on differential evolutionary algorithm. The simulation results show that DE with pseudo gradient can obtain better performance overall in comparison with classical DE variants. The pseudo gradient based DE with adaptive parameter section is compared with the existing adaptive DE algorithms. Also, the control parameter, step size are investigated to understand the mechanism of pseudo gradient in detail. }} @InProceedings{Li:2014:CECl, title = {Protein Folding Estimation Using Paired-Bacteria Optimizer}, author = {Mengshi Li and Tianyao Ji and Peter Wu and Shan He and Qinghua Wu}, pages = {2027--2032}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial Optimisation}, abstract = { Protein folding estimation attracts a large attention in the area of computational biology, due to its benefits on medical research and the challenge of NP-hard objective functions. In order to simulate the protein folding procedure and estimate the structure of the protein after folding, this paper adopts a Paired-Bacteria Optimiser (PBO), which is a biologically inspired Optimisation algorithm. Compared with most Evolutionary Algorithms (EAs), the computational complexity of PBO is much less. Therefore, it is suitable to be applied to solve NP-hard problem. The experimental studies is performed on several benchmark lattice protein combination. The experimental results demonstrated that PBO is able to estimate the folded protein structure with a superior convergence. }} @InProceedings{Zheng:2014:CECd, title = {A Self-Adaptive Group Search Optimizer with Elitist Strategy}, author = {Xiang-wei Zheng and Dian-jie Lu and Zhen-hua Chen}, pages = {2033--2039}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Coevolution and collective behaviour, Self-adaptation in evolutionary computation}, abstract = { To deal with the disadvantages of Group Search Optimiser (GSO) as slow convergence, easy entrapment in local optima and failure to use history information, a Self-adaptive Group Search Optimiser with Elitist strategy (SEGSO) is proposed in this paper. To maintain the group diversity, SEGSO employs a self-adaptive role assignment strategy, which determines whether a member is a scrounger or a ranger based on ConK consecutive iterations of the producer. On the other hand, scroungers are updated with elitist strategy based on simulated annealing by using history information to improve convergence and guarantee SEGSO to remain global search. Experimental results demonstrate that SEGSO outperform particle swarm optimiser and original GSO in convergence rate and escaping from local optima. }} @InProceedings{Xu:2014:CECc, title = {Optimization Based on Adaptive Hinging Hyperplanes and Genetic Algorithm}, author = {Jun Xu and Xiangming Xi and Shuning Wang}, pages = {2040--2046}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical Optimisation, Meta-modelling and surrogate models}, abstract = { This paper describes an Optimisation strategy based on the model of adaptive hinging hyperplanes (AHH) and genetic algorithm (GA). The sample points of physical model are approximated by the AHH model, and the resulting model is minimised using a modified GA. In the modified GA, each chromosome corresponds to a local optimum. A criterion based on \${$\backslash$}gamma\$-valid cut is used to judge whether the global optimum is reached. Simulation results show that if the parameters are carefully chosen, the global optimum of AHH minimisation is close to the optimum of the original function. }} @InProceedings{Zhu:2014:CEC, title = {Combining Multipopulation Evolutionary Algorithms with Memory for Dynamic Optimization Problems}, author = {Tao Zhu and Wenjian Luo and Lihua Yue}, pages = {2047--2054}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments}, abstract = { Both multipopulation and memory are widely used approaches in the field of evolutionary dynamic Optimisation. It would be interesting to examine the effect of the combinations of multipopulation algorithms (MPAs) and memory schemes. However, since most of the existing memory schemes are proposed with single population algorithms, straightforwardly applying them to MPAs may cause problems. By addressing the possible problems, a new memory scheme is proposed for MPAs in this paper. In the experiments, several existing memory schemes and the newly proposed scheme are combined with a MPA, i.e. the Species-based Particle Swarm Optimiser (SPSO), and these combinations are tested on cyclic and acyclic problems. The experimental results indicate that 1) straightforwardly using the existing memory schemes sometimes degrades the performance of SPSO even on cyclic problems; 2) the newly proposed memory scheme is very competitive. }} @InProceedings{Salehinejad:2014:CECa, title = {Micro-Differential Evolution with Vectorized Random Mutation Factor}, author = {Hojjat Salehinejad and Shahryar Rahnamayan and Hamid R. Tizhoosh}, pages = {2055--2062}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution}, abstract = { One of the main disadvantages of population-based evolutionary algorithms (EAs) is their high computational cost due to the nature of evaluation, specially when the population size is large. The micro-algorithms employ a very small number of individuals, which can accelerate the convergence speed of algorithms dramatically, while it highly increases the stagnation risk. One approach to overcome the stagnation problem can be increasing the diversity of the population. To do so, a micro-differential evolution with vectorised random mutation factor (MDEVM) algorithm is proposed in this paper, which uses the small size population benefit while preventing stagnation through diversification of the population. The proposed algorithm is tested on the 28 benchmark functions provided at the IEEE congress on evolutionary computation 2013 (CEC-2013). Simulation results on the benchmark functions demonstrate that the proposed algorithm improves the convergence speed of its parent algorithm. }} @InProceedings{Gao:2014:CECa, title = {Application of {BPSO} with {GA} in Model-Based Fault Diagnosis of Traction Substation}, author = {Song Gao and Zhigang Liu and Chenxi Dai and Xiao Geng}, pages = {2063--2069}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Particle swarm Optimisation (PSO), Engineering applications}, abstract = { In this paper, a hybrid evolutionary algorithm based on Binary Particle Swarm Optimisation (BPSO) and Genetic Algorithm (GA) is proposed to compute the minimal hitting sets in model-based diagnosis. And a minimal assurance strategy is proposed to ensure that the final output of algorithm is the minimal hitting sets. In addition, the logistic mapping of chaos theory is adopted to avoid the local optimum. The high efficiency of new algorithm is proved through comparing with other algorithms for different problem scales. Additionally, the new algorithm with logistic mapping could improve the realisation rate to almost 100\% from 96\%. At last, the new algorithm is used in the model-based fault diagnosis of traction substation. The results show that the new algorithm makes full use of the advantages of GA and BPSO and finds all the minimal hitting sets in 0.2369s, which largely meet the real-time requirement of fault diagnosis in the traction substation. }} @InProceedings{Du:2014:CEC, title = {Performance of {AI} Algorithms for Mining Meaningful Roles}, author = {Xuanni Du and Xiaolin Chang}, pages = {2070--2076}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Genetic algorithms, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Role-based access control (RBAC) is being today's dominant access control model due to its potential to mitigate the complexity and cost of access control administration. However, the migration from the access control lists (ACL) to RBAC for a large administration system may consume significant efforts, which challenges the adoption of RBAC. Role mining algorithms can significantly reduce the migration cost by providing a partially automatic construction of an RBAC policy. This paper explores Artificial Intelligence (AI) techniques in designing role mining algorithms, which can optimise policy quality in terms of policy size, user-attribute-based interpretability of the roles, and the combination of size and interpretability. We propose two algorithms, genetic algorithm (GA)-based and ant colony Optimisation (ACO)-based. GA-based algorithm works by starting with a set of all candidate roles and repeatedly removing roles. ACO-based algorithm works by starting with an empty policy and repeatedly adding candidate roles. We carry out extensive experiments with publicly available access control policies. The simulation results indicate that (1) the proposed algorithms achieves better performance than the corresponding existing algorithms. (2) GA-based approach produces better results than ACO-based approach. }} @InProceedings{Li:2014:CECm, title = {Using Estimation of Distribution Algorithm to Coordinate Decentralized Learning Automata for Meta-Task Scheduling}, author = {Jie Li and Junqi Zhang}, pages = {2077--2084}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Estimation of distribution algorithms, Engineering applications, Discrete and combinatorial Optimisation}, abstract = { Learning automaton (LA) is a reinforcement learning model that aims to determine the optimal action out of a set of actions. It is characterised by updating a selection probability set through a sequence of repetitive feedback cycles interacting with an environment. Decentralised learning automata (DLAs) consist of many learning automata (LAs) that learn at the same time. Each LA independently selects an action based on its own selection probability set. In order to provide an appropriate central coordination mechanism in DLAs, this paper proposes a novel decentralised coordination learning automaton (DCLA) using a new selection probability set which is combined with the probability sets derived from both LA and estimation of distribution algorithm (EDA). LA contributes to the own learning experience of each LA while EDA estimates the distribution of the whole swarm's promising individuals. Thus, decentralised LAs can be coordinated by EDA using the swarm's comprehensive knowledge. The proposed automaton is applied to solve the real problem of meta-task scheduling in heterogeneous computing system. Extensive experiments demonstrate a superiority of DCLA over other counterpart algorithms. The results show that the proposed DCLA provides an effective and efficient way to coordinate LAs for solving complicated problems. }} @InProceedings{Chatbri:2014:CEC, title = {A Modular Approach for Query Spotting in Document Images and Its Optimization Using Genetic Algorithms}, author = {Houssem Chatbri and Paul Kwan and Keisuke Kameyama}, pages = {2085--2092}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms}, abstract = { Query spotting in document images is a subclass of Content-Based Image Retrieval (CBIR) algorithms concerned with detecting occurrences of a query in a document image. Due to noise and complexity of document images, spotting can be a challenging task and easily prone to false positives and partially incorrect matches, thereby reducing the overall precision of the algorithm. A robust and accurate spotting algorithm is essential to our current research on sketch-based retrieval of digitised lecture materials. We have recently proposed a modular spotting algorithm in [Chatbri et al., 2014]. Compared to existing methods, our algorithm is both application-independent and segmentation-free. However, it faces the same challenges of noise and complexity of images. In this paper, inspired by our earlier research on optimising parameter settings for CBIR using an evolutionary algorithm [Kameyama et al., 2006][Okayama et al., 2008], we introduce a Genetic Algorithm-based Optimisation step in our spotting algorithm to improve each spotting result. Experiments using an image dataset of journal pages reveal promising performance, in that the precision is significantly improved but without compromising the recall of the overall spotting result. }} @InProceedings{Zhu:2014:CECa, title = {An Improved Genetic Algorithm for Dynamic Shortest Path Problems}, author = {Xuezhi Zhu and Wenjian Luo and Tao Zhu}, pages = {2093--2100}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments, Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { The shortest path (SP) problems are conventional combinatorial Optimisation problems. There are many deterministic algorithms solving the shortest path problems in static topologies. However, in dynamic topologies, these deterministic algorithms are not efficient due to the necessity of restart, while genetic algorithms (GAs) are good at solving dynamic Optimisation problems (DOPs). In this paper, an improved GA with four local search operators for dynamic shortest path (DSP) problems is proposed. The local search operators are inspired by Dijkstra's Algorithm and carried out when the topology changes to generate local shortest path trees, which are used to enhance the performance of the individuals in the population. The experimental results show that the proposed algorithm could obtain the solutions which adapt to new environments rapidly and produce high-quality solutions after environmental changes. }} @InProceedings{Wu:2014:CECd, title = {A Novel Genetic Algorithm Considering Measures and Phrases for Generating Melody}, author = {Chia-Lin Wu and Chien-Hung Liu and Chuan-Kang Ting}, pages = {2101--2107}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Music, Art, and Creativity}, abstract = { Composing music through evolutionary algorithms has received increasing attention recently. To establish a standard of composing, some studies were proposed on the basis of analysis on musicians, statistics of music details, and rule of thumbs. These methods have achieved some promising results; however, generating melody is still a formidable challenge to computer composition because of the considerable permutations of notes. This study develops a genetic algorithm (GA) based on music theory to generate melody. In particular, we use the rhythm of existing songs as the basis to generate new compositions instead of generating music from scratch; that is, the GA keeps the rhythm of an existing song and rearranges the pitches of all notes for a new composition. Three crossover operators are further proposed to improve the performance of GA on composition. The experimental results show that the GA can achieve satisfactory compositions. The three crossover operators outperform 2-point crossover in the fitness of resultant compositions. }} @InProceedings{Shi:2014:CEC, title = {Optimal Sizing of {DGs} and Storage for Microgrid with Interruptible Load Using Improved {NSGA-II}}, author = {Zhe Shi and Yonggang Peng and Wei Wei}, pages = {2108--2115}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Engineering applications}, abstract = { The rapid development of distributed generation (DG) has deeply transferred the power style. Microgrid is developed for better absorption of distributed generation and has been researched in recent years. Incorruptible load (IL) is another method to absorb the randomness and waviness of wind and solar energy, and is considered in this paper for more reliable and efficient deployment of DGs and storage in microgrid. A multi-objective optimisation model is proposed for microgrid power sources construction with distributed generation, storage and interruptible load. Objectives of the model are economic cost, environmental cost and annual interruption duration. The model is solved by employing improved NSGA-II with the input of temperature, light intensity, wind speed, and load curve. The case study shows that the Pareto optimal front which covers the optimal solutions under different circumstances is effectively obtained. Thus the supervisor can select the final scheme with full consideration of different objectives. The impacts of IL on economic and environmental cost are also analysed and demonstrated with many aspects. }} @InProceedings{B.-R.:2014:CEC, title = {Lion Algorithm for Standard and Large Scale Bilinear System Identification: A Global Optimization Based on Lion's Social Behavior}, author = {Rajakumar B. R.}, pages = {2116--2123}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical Optimisation, Large-scale problems, Engineering applications}, abstract = { Nonlinear system identification process, especially bilinear system identification process exploits global Optimisation algorithms for betterment of identification precision. This paper attempts to introduce a new Optimisation algorithm called as Lion algorithm to accomplish the system characteristics precisely. Our algorithm is a simulation model of the lion's unique characteristics such as territorial defence, territorial takeover, laggardness exploitation and pride. Experiments are conducted by identifying a nonlinear rationale digital benchmark system using standard bilinear model and comparisons are made with prominent genetic algorithm and differential evolution. Subsequently, curse of dimensionality is also experimented by defining a large scale bilinear model, i.e. bilinear system with 1023 bilinear kernel models, to identify the same digital benchmark system. Lion algorithm dominates when using standard bilinear model, whereas it is equivalent to differential evolution and better than genetic algorithm when using large scale bilinear model. }} @InProceedings{Wang:2014:CECd, title = {Intelligent Search Optimized Edge Potential Function ({EPF}) Approach to Synthetic Aperture Radar ({SAR}) Scene Matching}, author = {Yifei Wang and Jihao Yin}, pages = {2124--2131}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Engineering applications, Intelligent systems applications, Evolutionary Computer Vision }, abstract = { Research on synthetic aperture radar (SAR) scene matching in the aircraft end guidance has a significant value for both research and real-world application. The conventional scene matching methods, however, suffer many disadvantages such as heavy computation burden and low convergence rate so that these methods cannot meet the requirement of end-guidance system in terms of fast and real-time data processing. Furthermore, there are complex noises in the SAR image, which also compromise the effectiveness of using the conventional scene matching methods. To address the above issues, in this paper, the intelligent Optimisation method, Free Search with Adaptive Differential Evolution Exploitation and Quantum Inspired Exploration, has been introduced to tackle the SAR scene matching problem. We first establish the effective similarity measurement function for target edge feature matching through introducing the edge potential function (EPF) model. Then, a new method, ADEQFS-EPF, has been proposed for SAR scene matching. In ADEQFS-EPF, the previous studied theoretical model, ADEQFS, is combined with EPF model. We also employed three recent proposed evolutionary algorithms to compare against the proposed method on optical and SAR datasets. The experiments based on Matlab simulation have verified the effectiveness of the application of ADEQFS and EPF model to the field of SAR scene matching. }} % Session: ThE2-1 Multi-Objective Evolutionary Algorithms II @InProceedings{Wang:2014:CECe, title = {A Replacement Strategy for Balancing Convergence and Diversity in {MOEA/D}}, author = {Zhenkun Wang and Qingfu Zhang and Maoguo Gong and Aimin Zhou}, pages = {2132--2139}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making, Multi-objective evolutionary algorithms, Multiobjective Optimisation}, abstract = { This paper studies the replacement schemes in MOEA/D and proposes a new replacement named global replacement. It can improve the performance of MOEA/D. Moreover, trade-offs between convergence and diversity can be easily controlled in this replacement strategy. It shows that different problems need different trade-offs between convergence and diversity. We test the MOEA/D with this global replacement on three sets of benchmark problems to demonstrate its effectiveness. }} @InProceedings{Li:2014:CECn, title = {A Test Problem for Visual Investigation of High-Dimensional Multi-Objective Search}, author = {Miqing Li and Shengxiang Yang and Xiaohui Liu}, pages = {2140--2147}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { An inherent problem in multiobjective Optimisation is that the visual observation of solution vectors with four or more objectives is infeasible, which brings major difficulties for algorithmic design, examination, and development. This paper presents a test problem, called the Rectangle problem, to aid the visual investigation of high-dimensional multiobjective search. Key features of the Rectangle problem are that the Pareto optimal solutions 1) lie in a rectangle in the two-variable decision space and 2) are similar (in the sense of Euclidean geometry) to their images in the four-dimensional objective space. In this case, it is easy to examine the behaviour of objective vectors in terms of both convergence and diversity, by observing their proximity to the optimal rectangle and their distribution in the rectangle, respectively, in the decision space. Fifteen algorithms are investigated. Under performance of Pareto-based algorithms as well as most state-of-the-art many-objective algorithms indicates that the proposed problem not only is a good tool to help visually understand the behaviour of multiobjective search in a high-dimensional objective space but also can be used as a challenging benchmark function to test algorithms' ability in balancing the convergence and diversity of solutions. }} @InProceedings{Menchaca-Mendez:2014:CEC, title = {{MD-MOEA} : A New {MOEA} Based on the Maximin Fitness Function and {Euclidean} Distances between Solutions}, author = {Adriana Menchaca-Mendez and Carlos A. Coello Coello}, pages = {2148--2155}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Multiobjective Optimisation}, abstract = { In this paper, we propose a new selection mechanism based on the maxi-min fitness function and a technique based on Euclidean distances between solutions to improve the diversity of the population in objective function space. Our new selection mechanism is incorporated into a multi-objective evolutionary algorithm (MOEA) which uses the operators of NSGA-II (crossover and mutation) to generate new individuals, giving rise to the so-called "Maximin-Distances Multi-Objective Evolutionary Algorithm (MD-MOEA)". Our MD-MOEA is validated using standard test functions taken from the specialised literature, having three to six objective functions. MD-MOEA is compared with respect to MC-MOEA (which is based on the maximin fitness function and a clustering technique), MOEA/D using Penalty Boundary Intersection (PBI), which is based on decomposition, and SMS-EMOA-HYPE (a version of SMS-EMOA that uses a fitness assignment based on the use of an approximation of the hypervolume indicator). Our preliminary results indicate that our MD-MOEA is a good alternative to solve multi-objective Optimisation problems having both low dimensionality and high dimensionality in objective function space because it obtains better results than MC-MOEA and MOEA/D in most cases and it is competitive with respect to SMS-EMOA-HYPE (in fact, it outperforms SMS-EMOA-HYPE in problems of high dimensionality) but at a much lower computational cost. }} @InProceedings{Li:2014:CECo, title = {Multiobjective Test Problems with Complicated {Pareto} Fronts: Difficulties in Degeneracy}, author = {Hui Li and Qingfu Zhang and Jingda Deng}, pages = {2156--2163}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms}, abstract = { It is well-established that the shapes of Pareto-optimal fronts (POFs) can affect the performance of some multiobjective Optimisation methods. The most well-known characteristics on the shape of POFs are convexity and discontinuity. In this paper, we investigate the construction of multiobjective test problems with complicated POFs, of which its local parts could have mixed dimensionalities. For example, in the case of 3 objectives, some parts of POFs can be 1-D curves while others could be 2-D surfaces. We formulate eight test problems, called CPFT1-8, with such a feature. To study the difficulties of these test problems, we conducted some experiments with two state-of-the-art algorithms MOEA/D and NSGA-II, and analysed their performances. }} @InProceedings{Souza:2014:CECa, title = {A Comparison Study of Binary Multi-Objective Particle Swarm Optimization Approaches for Test Case Selection}, author = {Luciano Souza and Ricardo Prudencio and Flavia Barros}, pages = {2164--2171}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Multi-objective evolutionary algorithms, Particle swarm Optimisation (PSO), Real-world applications}, abstract = { During the software testing process many test suites can be generated in order to evaluate and assure the quality of the products. In some cases the execution of all suites can not fit the available resources (time, people, etc). Hence, automatic Test Case (TC) selection could be used to reduce the suites based on some selection criterion. This process can be treated as an Optimisation problem, aiming to find a subset of TCs which optimises one or more objective functions (i.e., selection criteria). The majority of search-based works focus on single-objective selection. In this light, we developed mechanisms for functional TC selection which considers two objectives simultaneously: maximise requirements' coverage while minimising cost in terms of TC execution effort. These mechanisms were implemented by deploying multi-objective techniques based on Particle Swarm Optimisation (PSO). Due to the drawbacks of original binary version of PSO we implemented five binary PSO algorithms and combined them with a multi-objective versions of PSO in order to create new Optimisation strategies applied to TC selection. The experiments were performed on two real test suites, revealing the feasibility of the proposed strategies and the differences among them. }} @InProceedings{Pilat:2014:CEC, title = {The Effect of Different Local Search Algorithms on the Performance of Multi-Objective Optimizers}, author = {Martin Pilat and Roman Neruda}, pages = {2172--2179}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Heuristics, metaheuristics and hyper-heuristics}, abstract = { Several contemporary multi-objective surrogate-based algorithms use some kind of local search operator. The search technique used in this operator can largely affect the performance of the multi-objective optimiser as a whole, however, little attention is often paid to the selection of this technique. In this paper, we compare three different local search techniques and evaluate their effect on the performance of two different surrogate based multi-objective optimisers. The algorithms are evaluated using the well known ZDT and WFG benchmark suites and recommendations are made based on the results. }} % Session: ThE2-2 Cultural Algorithms and Knowledge Extraction in Evolutionary Algorithms @InProceedings{Ali:2014:CEC, title = {Cultural Algorithms Applied to the Evolution of Robotic Soccer Team Tactics: A Novel Perspective}, author = {Mostafa Ali and Abdulmalik Morghem and Jafar AlBadarneh and Rami Al-Gharaibeh and Ponnuthurai Suganthan and Robert Reynolds}, pages = {2180--2187}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Cultural algorithms, Cultural Algorithms: Theory and Applications}, abstract = { Cultural Algorithms have been previously employed to model the emergence of cooperative behaviours of agents in different multi-agent systems. In this paper, a simplified and adaptive version will be used as the basis to generate cooperative behaviours within a team of soccer players using different team formations and effective plays. This system can be used as a tutorial for the application of Cultural Algorithms for the coordination of groups of agents in complex multi-agent dynamic environments. Simplified Cultural Algorithms were successful in effectively learning different types of plays, including active and passive protagonists, within a small number of generations. Successful learning includes the coordination of adjustments of the team members to develop the most suitable team formations for every scenario. Experimental results enable us to conclude that Cultural Algorithms, when configured properly, in order to produce significant results, can perform very competitively when compared to other types of learning strategies and case-based game plays. }} @InProceedings{Juan:2014:CEC, title = {Cultural Learning for Multi-Agent System and Its Application to Fault Management}, author = {Teran Juan and Aguilar Jose and Cerrada Mariela}, pages = {2188--2195}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Collaborative Learning and Optimisation, Cultural algorithms, Coevolution and collective behaviour}, abstract = { It is usually agreed that a system capable of learning deserves to be called intelligent; and conversely, a system being considered as intelligent is, among other things, usually expected to be able to learn. Learning always has to do with the self-improvement of future behaviour based on past experience. In this paper we present a learning model for Multi-Agent System, which aims to the Optimisation of coordination schemes through a collective learning process based on Cultural Algorithms. }} @InProceedings{Stanley:2014:CEC, title = {Analyzing Prehistoric Hunter Behavior with Cultural Algorithms}, author = {Samuel Stanley and Thomas Palazzolo and David Warnke}, pages = {2196--2205}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Cultural Algorithms: Theory and Applications}, abstract = { This paper details a cultural algorithm (CA) system designed to assist archaeological expedition teams in the task of finding historic artifacts. In our system, the goals that the agents are trying to achieve continuously change as the environment changes. We are thus able to simulate the real-world challenge of a dynamic environment that human cultures must deal with and react to, making our system a very useful tool for finding the archaeological remains of such cultures. Although it is very new, our system has already had yielded promising results in the service of Dr. John O'Shea's Lake Huron expedition team which is studying the prehistoric Alpena-Amberley Land Bridge. We hope to use it to assist other expeditions as well in the near future. }} @InProceedings{Judeh:2014:CEC, title = {{GSCA}: Reconstructing Biological Pathway Topologies Using a Cultural Algorithms Approach}, author = {Thair Judeh and Thaer Jayyousi and Lipi Acharya and Robert Reynolds and Dongxiao Zhu}, pages = {2206--2213}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Cultural algorithms, Cultural Algorithms: Theory and Applications}, abstract = { With the increasing availability of gene sets and pathway resources, novel approaches that combine both resources to reconstruct networks from gene sets are of interest. Currently, few computational approaches explore the search space of candidate networks using a parallel search. In particular, search agents employed by evolutionary computational approaches may better escape false peaks compared to previous approaches. It may also be thought that gene sets may model signal transduction events, which refer to linear chains or cascades of reactions starting at the cell membrane and ending at the cell nucleus. These events may be indirectly observed as a set of unordered and overlapping gene sets. Thus, the goal is to reverse engineer the order information within each gene set to reconstruct the underlying source network using prior knowledge to limit the search space. We propose the Gene Set Cultural Algorithm (GSCA) to reconstruct networks from unordered gene sets. We introduce a robust heuristic based on the arborescence of a directed graph that performs well for random topological sort orderings across gene sets simulated for four E. coli networks and five In-silico networks from the DREAM3 and DREAM4 initiatives, respectively. Furthermore, GSCA performs favourably when reconstructing networks from randomly ordered gene sets for the aforementioned networks. Finally, we note that from a set of 23 gene sets discredited from a set of 300 S. cerevisiae expression profiles, GSCA reconstructs a network preserving most of the weak order information found in the KEGG Cell Cycle pathway, which was used as prior knowledge. }} @InProceedings{Che:2014:CEC, title = {A Social Metrics Based Process Model on Complex Social System}, author = {Xiangdong Che and Robert Reynolds}, pages = {2214--2221}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Cultural algorithms, Cultural Algorithms: Theory and Applications}, abstract = { In previous work, we investigated the performance of Cultural Algorithms (CA) over the complete range of system complexities in a benchmarked environment. In this paper the goal is to discover whether there is a similar internal process going on in CA problem solving, regardless of the complexity of the problem. We are to monitor the "vital signs" of a cultural system during the problem solving process to determine whether it was on track or not and infer the complexity class of a social system based on its "vital signs". We first demonstrate how the learning curve for a Cultural System is supported by the interaction of the knowledge sources. Next a circulatory system metaphor is used to describe how the exploratory knowledge sources generate new information that is distributed to the agents via the Social Fabric network. We then conclude that the Social Metrics are able to indicate the progress of the problem solving in terms of its ability to periodically lower the innovation cost for the performance of a knowledge source which allows the influenced population to expand and explore new solution possibilities as seen in the dispersion metric. Hence we present the possibility to assess the complexity of a system's environment by looking at the Social Metrics. }} @InProceedings{Zhang:2014:CECh, title = {Online Knowledge-Based Evolutionary Multi-Objective Optimization}, author = {Bin Zhang and Kamran Shafi and Hussein Abbass}, pages = {2222--2229}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms}, abstract = { Knowledge extraction from a multi-objective Optimisation process has important implications including a better understanding of the Optimisation process and the relationship between decision variables. The extant approaches, in this respect, rely on processing the post-Optimisation Pareto sets for automatic rule discovery using statistical or machine learning methods. However such approaches fall short of providing any information during the progress of the Optimisation process, which can be critical for decision analysis especially if the problem is dynamic. In this paper, we present a multi-objective Optimisation framework that uses a knowledge-based representation to search for patterns of Pareto optimal design variables instead of conventional point form solution search. The framework facilitates the online discovery of knowledge during the Optimisation process in the form of interpretable rules. The core contributing idea of our research is that we apply multi-objective evolutionary process on a population of bounding hypervolumes, or rules, instead of evolving individual point-based solutions. The framework is generic in a sense that any existing multi-objective Optimisation algorithm can be adapted to evaluate the rule quality based on the sampled solutions from the bounded space. An instantiation of the framework using hyperrectangular representation and nondominated sorting based rule evaluation is presented in this paper. Experimental results on a specifically designed test function as well as some standard test functions are presented to demonstrate the working and convergence properties of our algorithm. }} % Special Session: ThE2-3 Single Objective Numerical Optimisation III @InProceedings{Polakova:2014:CEC, title = {Controlled Restart in Differential Evolution Applied to {CEC2014} Benchmark Functions}, author = {Radka Polakova and Josef Tvrdik and Petr Bujok}, pages = {2230--2236}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { A controlled restart in differential evolution (DE) is proposed. The conditions of restart are derived from the difference of maximum and minimum values of the objective function and the estimated maximum distance among the points in the current population. The restart is applied in a competitive-adaptation variant of DE. This DE algorithm with the controlled restart is used in the solution of the benchmark problems defined for the CEC 2014 competition. Two control parameters of restart are set up intuitively. The population size, which is the only control parameter of competitive-adaptation variant of DE, is set up to the values based on a short preliminary experimentation. }} @InProceedings{Dhebar:2014:CEC, title = {Non-Uniform Mapping in Real-Coded Genetic Algorithms}, author = {Yashesh Dhebar and Kalyanmoy Deb and Sunith Bandaru}, pages = {2237--2244}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Genetic algorithms}, abstract = { Genetic algorithms have been used as Optimisation tool using evolutionary strategies. Genetic algorithms cover three basic steps for population refinement selection, cross-over and mutation. In normal Real-coded genetic algorithm(RGA), the population of real variables generated after population refinement operations, is used as it is for the computation of the objective function. In this paper we have shown the effect made by mapping the refined population towards better solutions and thereby creating more biased search. The mapping used was non-uniform in nature and was the function of the position of the individual w.r.t. the best solution obtained so far in the algorithm, and hence the name Non-Uniform RGA or in short NRGA. Tests were performed on standard benchmark problems. The results were promising and should encourage further research in this dimension. }} @InProceedings{Philippe:2014:CEC, title = {Bandits Attack Function Optimization}, author = {Preux Philippe and Munos Remi and Valko Michal}, pages = {2245--2252}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { We consider function Optimisation as a sequential decision making problem under the budget constraint. Such constraint limits the number of objective function evaluations allowed during the Optimisation. We consider an algorithm inspired by a continuous version of a multi-armed bandit problem which attacks this Optimisation problem by solving the trade off between exploration (initial quasi-uniform search of the domain) and exploitation (local Optimisation around the potentially global maxima). We introduce the so-called Simultaneous Optimistic Optimisation (SOO), a deterministic algorithm that works by domain partitioning. The benefit of such an approach are the guarantees on the returned solution and the numerical efficiency of the algorithm. We present this machine learning rooted approach to Optimisation, and provide the empirical assessment of SOO on the CEC'2014 competition on single objective real-parameter numerical Optimisation test-suite. }} @InProceedings{Bujok:2014:CEC, title = {Differential Evolution with Rotation-Invariant Mutation and Competing-Strategies Adaptation}, author = {Petr Bujok and Josef Tvrdik and Radka Polakova}, pages = {2253--2258}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { A new variant of the adaptive differential evolution algorithm was proposed and tested experimentally on the CEC 2014 test suite. In the new variant, the adaptation is based on the competition of several strategies. A part of strategies in the pool uses the rotation-invariant current-to-pbest mutation in the novel algorithm. The aim of the experimental comparison was to find whether the presence of the rotation-invariant strategy is able to improve the efficiency of the differential evolution algorithm, especially in problems with rotated objective functions. The results of the experiments showed that the new variant performed well in a few of the test problems, while no apparent benefit was observed in the majority of the benchmark problems. }} @InProceedings{Hu:2014:CECb, title = {Partial Opposition-Based Adaptive Differential Evolution Algorithms: Evaluation on the {CEC 2014} Benchmark Set for Real-Parameter Optimization}, author = {Zhongyi Hu and Yukun Bao and Tao Xiong}, pages = {2259--2265}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Differential evolution}, abstract = { Opposition-based Learning (OBL) has been reported with an increased performance in enhancing various Optimisation approaches. Instead of investigating the opposite point of a candidate in OBL, this study proposed a partial opposition-based learning (POBL) schema that focuses a set of partial opposite points (or partial opposite population) of an estimate. Furthermore, a POBL-based adaptive differential evolution algorithm (POBL-ADE) is proposed to improve the effectiveness of ADE. The proposed algorithm is evaluated on the CEC2014's test suite in the special session and competition for real parameter single objective Optimisation in IEEE CEC 2014. Simulation results over the benchmark functions demonstrate the effectiveness and improvement of the POBL-ADE compared with ADE. }} @InProceedings{Liang:2014:CECb, title = {Memetic Differential Evolution Based on Fitness {Euclidean}-Distance Ratio}, author = {J. J. Liang and B. Y. Qu and H. Song and Z. G. Shang}, pages = {2266--2273}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { In this paper, a differential evolution algorithm based on fitness Euclidean distance ratio which was proposed to maintain multiple peaks in the multimodal Optimisation problems was modified to solve the complex single objective real parameter Optimisation problems. With the fitness Euclidean-distance ratio technique, the diversity of the population was kept to enhance the exploration ability. And in order to improve the exploitation ability, the Quasi-Newton method was combined. The performance of the proposed method on the set of benchmark functions provided by CEC2014 competition on single objective real parameter numerical Optimisation was reported. }} % Session: ThE2-4 Music, Art, Creativity, Games and Multi-Agent Systems @InProceedings{Campbell:2014:CEC, title = {A Self Organising Map Based Method for Understanding Features Associated with High Aesthetic Value Evolved Abstract Images}, author = {Allan Campbell and Vic Ciesielski and Karen Trist}, pages = {2274--2281}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Music, Art, and Creativity}, abstract = { We show a method that allows the pixel data of a set of images to be analysed independently of any set of computed features. If the high and low aesthetic value images can be separated in the high dimensional space of pixel intensities then for any given set of features computed from the images, those features relevant to high aesthetic value can be determined and the range of feature values that correlate with high aesthetic appeal can be quantified. The method uses the Self Organising Map to project raw pixel data of images onto a feature map. The aesthetic class of these images is overlaid on the feature map, yielding a semantic map. Average feature values are visualised in gray-scale heat maps and features relevant to aesthetic value are identified. We call this the Pixel Array Self Organising Map (PASOM) method. For the set of images analysed, brightness and texture features were identified as being discriminatory between images of high and low aesthetic value. High aesthetic value images tend to have higher brightness and richer textures. These findings were corroborated by a professional artist/photographer as being consistent with the principles for attaining aesthetic value in visual media. The PASOM method yields a semantic map and a visualisation of feature value variation that together make possible a detailed analysis of features associated with the aesthetic value of images. }} @InProceedings{Fernandez-de-Vega:2014:CEC, title = {When Artists Met {Evospace-i}}, author = {Francisco Fernandez de Vega and Mario Garcia-Valdez and Lilian Navarro and Cayetano Cruz and Patricia Hernandez and Tania Gallego and J. Vicente Albarran}, pages = {2282--2289}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Art and music, Interactive evolutionary computation, Evolutionary Computation for Music, Art, and Creativity}, abstract = { This paper presents a new step towards a hard goal: establishing a stronger collaboration between the art world and the field of Evolutionary Algorithms, so that both can benefit. Two were the main reasons for pursuing this goal: on the one hand the aim of studying human creative processes that may allow in the future improving computer based creativity; on the other hand we wanted to both improve available software tools and also propose a methodology allowing artists to develop collective evolutionary based artistic experiences. This paper focuses in the second goal, and shows a new addition to EvoSpace-i software tool, as well as to the methodology applied, and how it was employed by a team of artists when creating a new collective artwork. }} @InProceedings{Sephton:2014:CEC, title = {Parallelization of Information Set {Monte Carlo} Tree Search}, author = {Nicholas Sephton and Peter Cowling and Edward Powley and Daniel Whitehouse and Nicholas Slaven}, pages = {2290--2297}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence and Games}, abstract = { Process parallelisation is more important than ever, as most modern hardware contains multiple processors and advanced multi-threading capability. This paper presents an analysis of the parallel behaviour of Information Set Monte Carlo Tree Search and the Upper Confidence Bounds for Trees (UCT) variant of MCTS, and certain parallelisation techniques (specifically Tree Parallelisation) have different effects upon ISMCTS and Plain UCT. The paper presents a study of the relative effectiveness of different types of parallelisation, including Root, Tree, Tree with Virtual Loss, and Leaf. }} @InProceedings{Wang:2014:CECf, title = {Comparing Crossover Operators in Neuro-Evolution with Crowd Simulations}, author = {Sunrise Wang and James Gain and Geoff Nitschke}, pages = {2298--2305}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolved neural networks, Evolutionary games and multi-agent systems}, abstract = { Crowd simulations are a set techniques used to control groups of agents and are exemplified by scenes from movies such as The Lord of the Rings and Inception. A problem which all crowd simulation techniques suffer from is the balance between control of the crowd behaviour and the autonomy of the agents. One possible solution to this problem is to use Neuro-Evolution(NE) to evolve the agent models so that the agents behave realistically and the emergent crowd behaviour is controllable. Since this is not an application area which has been investigated much, it is unknown which NE parameters and operators work well. This paper attempts to address this by comparing the performance of a set of crossover operators and with a range of probabilities across various crowd simulations. Overall it was found that Laplace crossover worked the best across all our simulations. }} @InProceedings{Davila:2014:CEC, title = {Genotype Coding, Diversity, and Dynamic Environments: A Study on an Evolutionary Neural Network Multi-Agent System}, author = {Jaime Davila}, pages = {2306--2313}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolved neural networks, Dynamic and uncertain environments, Evolutionary games and multi-agent systems}, abstract = { This paper reports the effects that different coding schemes at the genetic level have on the evolution of neural network multi-agent systems that operate under dynamic (changing) environments. Types of NN encoding include direct encoding of weights and three different L-Systems. Empirical results show that even variations within the same type of coding scheme can have considerable effects on evolution. Several different analysis of both genotypes and phenotypes are used in order to explain the differences caused by the coding schemes. }} @InProceedings{Perez:2014:CEC, title = {The 2013 Multi-Objective Physical Travelling Salesman Problem Competition}, author = {Diego Perez and Edward Powley and Daniel Whitehouse and Spyridon Samothrakis and Simon Lucas and Peter Cowling}, pages = {2314--2321}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Games, Adaptive dynamic programming and reinforcement learning, Multiobjective Optimisation}, abstract = { This paper presents the game, framework, rules and results of the Multi-objective Physical Travelling Salesman Problem (MO-PTSP) Competition, that was held at the 2013 IEEE Conference on Computational Intelligence in Games (CIG). The MO-PTSP is a real-time game that can be seen as a modification of the Travelling Salesman Problem, where the player controls a ship that must visit a series of way-points in a maze while minimising three opposing goals: time spent, fuel consumed and damage taken. The rankings of the competition are computed using multi-objective concepts, a novel approach in the field of game artificial intelligence competitions. The winning entry of the contest is also explained in detail. This controller is based on the Monte Carlo Tree Search algorithm, and employed Covariance Matrix Adaptation Evolution Strategy (CMA-ES) for parameter tuning. }} % Session: ThE2-5 Real-World Applications I @InProceedings{Shao:2014:CEC, title = {Vessel Track Correlation and Association Using Fuzzy Logic and Echo State Networks}, author = {Hang Shao and Rami Abielmona and Rafael Falcon and Nathalie Japkowicz}, pages = {2322--2329}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence for Security, Surveillance and Defence (CISSD)}, abstract = { Tracking moving objects is a task of the utmost importance to the defence community. As this task requires high accuracy, rather than employing a single detector, it has become common to use multiple ones. In such cases, the tracks produced by these detectors need to be correlated (if they belong to the same sensing modality) or associated (if they were produced by different sensing modalities). In this work, we introduce Computational-Intelligence-based methods for correlating and associating various contacts and tracks pertaining to maritime vessels in an area of interest. Fuzzy k-Nearest Neighbours will be used to conduct track correlation and Fuzzy C-Means clustering will be applied for association. In that way, the uncertainty of the track correlation and association is handled through fuzzy logic. To better model the state of the moving target, the traditional Kalman Filter will be extended using an Echo State Network. Experimental results on five different types of sensing systems will be discussed to justify the choices made in the development of our approach. In particular, we will demonstrate the judiciousness of using Fuzzy k-Nearest Neighbours and Fuzzy C-Means on our tracking system and show how the extension of the traditional Kalman Filter by a recurrent neural network is superior to its extension by other methods. }} @InProceedings{Wang:2014:CECg, title = {Automatic Target Recognition Using Multiple-Aspect Sonar Images}, author = {Xiaoguang Wang and Xuan Liu and Nathalie Japkowicz and Stan Matwin}, pages = {2330--2337}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence for Security, Surveillance and Defence (CISSD)}, abstract = { Automatic Target Recognition (ATR) methods have been successfully applied to detect possible objects or regions of interest in sonar imagery. It is anticipated that the additional information obtained from additional views of an object should improve the classification performance over single-aspect classification. In this paper the detection of mine-like objects (MLO) on the seabed from multiple side scan sonar views is considered. We transform the multiple-aspect classification problem into a multiple-instance learning problem and present a framework based upon the concepts of multiple-instance classifiers. Moreover, we present another framework based upon the Dempster-Shafer (DS) concept of fusion from single-view classifiers. Our experimental results indicate that both the presented frameworks can be successfully used in mine-like object classification. }} @InProceedings{Yu:2014:CECg, title = {Base Station Switching Problem for Green Cellular Networks with Social Spider Algorithm}, author = {James J.Q. Yu and Victor O.K. Li}, pages = {2338--2344}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Evolutionary Computation for Planning and Scheduling}, abstract = { With the recent explosion in mobile data, the energy consumption and carbon footprint of the mobile communications industry is rapidly increasing. It is critical to develop more energy-efficient systems in order to reduce the potential harmful effects to the environment. One potential strategy is to switch off some of the under-used base stations during off-peak hours. In this paper, we propose a binary Social Spider Algorithm to give guidelines for selecting base stations to switch off. In our implementation, we use a penalty function to formulate the problem and manage to bypass the large number of constraints in the original Optimisation problem. We adopt several randomly generated cellular networks for simulation and the results indicate that our algorithm can generate superior performance. }} @InProceedings{Wang:2014:CECh, title = {Deployment Optimization of Near Space Airships Based on {MOEA/D} with Local Search}, author = {Zhao Wang and Maoguo Gong and Qing Cai and Lijia Ma and Licheng Jiao}, pages = {2345--2352}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic, multi-meme and hybrid algorithms, Memetic Computing}, abstract = { The near space communication system is a burgeoning communication system. This system has many advantages over the satellite and terrestrial networks. Being built on the near space airships, the deployment of the airships has a significant impact on the performance of the system. Various factors should be taken into consideration to build such a system of which some objectives relate with each other and specific areas weight objectives differently. The evolutionary multiobjective Optimisation can fulfil the purpose to provide a series of choices of the deployment scheme. In this paper, a model of such a system is proposed and solved by the multiobjective evolutionary algorithm based on decomposition. Cases with different number of airships are tested and the Pareto Front is obtained. In order to increase the density of the Pareto Front, a local search method based on the positions of the airships is proposed. The experiment shows that the local search method can effectively increase the number of Pareto solutions obtained. }} @InProceedings{Tung:2014:CEC, title = {Novel Traffic Signal Timing Adjustment Strategy Based on Genetic Algorithm}, author = {Hsiao-Yu Tung and Wei-Chiu Ma and Tian-Li Yu}, pages = {2353--2360}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Real-world applications, Engineering applications}, abstract = { Traffic signal timing Optimisation problem aims at alleviating traffic congestion and shortening the average traffic time. However, most existing research considered only the information of one or few intersections at a time. Those local Optimisation methods may experience a decrease in performance when facing large-scale traffic networks. In this paper, we propose a cellular automaton traffic simulation system and conduct tests on two different Optimisation schemes. We use Genetic Algorithm (GA) for global Optimisation and Expectation Maximisation (EM) as well as car flow for local Optimisation. Empirical results show that the GA method outperforms the EM method. Then, we use linear regression to learn from the global optimal solution obtained by GA and propose a new adjustment strategy that outperforms recent Optimisation methods. }} @InProceedings{Mauser:2014:CEC, title = {Encodings for Evolutionary Algorithms in Smart Buildings with Energy Management Systems}, author = {Ingo Mauser and Marita Dorscheid and Florian Allerding and Hartmut Schmeck}, pages = {2361--2366}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Representation and operators, Self-adaptation in evolutionary computation}, abstract = { In energy systems, the transition from traditional, centralised architecture and controllable generation to an ever more decentralised and volatile generation due to an increasing use of renewable energy sources arises new challenges for the management and balancing of the electricity grid. These can be met through energy management systems (EMS) that enable flexible consumption and production of energy on the demand side of the grid. The EMS for smart buildings that is used within this paper allows for the integration of a multitude of devices through an architectural approach which is similar to plug-and-play. These devices can then be optimised to a flexible load shape by an Evolutionary Algorithm (EA). The differentiated Optimisation capabilities of the devices require adequate encoding schemes. Such schemes are the major contribution of this paper. The aptitude of these encodings is shown and validated through the simulation of smart buildings with different configurations, both concerning quantitative and qualitative benefits to be achieved according to energy systems' transition and users' objectives. }} % Session: FrE1-1 Differential Evolution @InProceedings{Mayo:2014:CEC, title = {Evolving Artificial Datasets to Improve Interpretable Classifiers}, author = {Michael Mayo and Quan Sun}, pages = {2367--2374}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Data mining, Classification, clustering and data analysis}, abstract = { Differential Evolution can be used to construct effective and compact artificial training datasets for machine learning algorithms. In this paper, a series of comparative experiments are performed in which two simple and interpretable supervised classifiers (specifically, Naive Bayes and linear Support Vector Machines) are trained (i) directly on "real" data, as would be the normal case, and (ii) indirectly, using special artificial datasets derived from real data via evolutionary Optimisation. The results across several challenging test problems show that supervised classifiers trained indirectly using our novel evolution based approach produce models with superior predictive classification performance. Besides presenting the accuracy of the learnt models, we also analyse the sensitivity of our artificial data Optimisation process to Differential Evolution's parameters, and then we examine the statistical characteristics of the artificial data that is evolved. }} @InProceedings{Varela:2014:CEC, title = {Differential Evolution in Constrained Sampling Problems}, author = {Gervasio Varela and Pilar Caamano and Felix Orjales and Alvaro Deibe and Fernando Lopez-Pena and Richard Duro}, pages = {2375--2382}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Real-world applications, Constraint handling}, abstract = { This work proposes a set of modifications to the Differential Evolution algorithm in order to make it more efficient in solving a particular category of problems, the so called Constrained Sampling problems. In this type of problems, which are usually related to the on-line real-world application of evolution, it is not always straightforward to evaluate the fitness landscapes due to the computational cost it implies or to physical constraints of the specific application. The fact is that the sampling or evaluation of the offspring points within the fitness landscape generally requires a decoding phase that implies physical changes over the parents or elements used for sampling the landscape, whether through some type of physical migration from their locations or through changes in their configurations. Here we propose a series of modifications to the Differential Evolution algorithm in order to improve its efficiency when applied to this type of problems. The approach is compared to a standard DE using some common real-coded benchmark functions and then it is applied to a real constrained sampling problem through a series of real experiments where a set of Unmanned Aerials Vehicles is used to find shipwrecked people. }} @InProceedings{Plagianakos:2014:CEC, title = {Unsupervised Clustering and Multi-Optima Evolutionary Search}, author = {Vassilis Plagianakos}, pages = {2383--2390}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution}, abstract = { This paper pursues a course of investigation of an approach to combine Evolutionary Computation and Data Mining for the location and computation of multiple local and global optima of an objective function. To accomplish this task we exploit the spatial concentration of the population members around the optima of the objective function. Such concentration regions are determined by applying clustering algorithms on the actual positions of the members of the population. Subsequently, the evolutionary search is confined in the interior of the regions discovered. To enable the simultaneous discovery of more than one global and local optima, we propose the use of clustering algorithms that also provide intuitive approximations for the number of clusters. Furthermore, the proposed scheme has often the potential of accelerating the convergence speed of the Evolutionary Algorithm, without the need for extra function evaluations. }} @InProceedings{Qiu:2014:CEC, title = {A Novel Differential Evolution ({DE}) Algorithm for Multi-Objective Optimization}, author = {Xin Qiu and Jianxin Xu and Kay Chen Tan}, pages = {2391--2396}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Multi-objective evolutionary algorithms}, abstract = { Convergence speed and parametric sensitivity are two issues that tend to be neglected when extending Differential Evolution (DE) for multi-objective Optimisation. To fill in this gap, we propose a multi-objective DE variant with an extraordinary mutation strategy and unfixed parameters. Wise trade off between convergence and diversity is achieved via the novel cross-generation mutation operators. In addition, a dynamic mechanism enables the parameters to evolve continuously during the Optimisation process. Empirical results show that the proposed algorithm is powerful in handling multi-objective problems. }} @InProceedings{St-Pierre:2014:CEC, title = {Differential Evolution Algorithm Applied to Non-Stationary Bandit Problem}, author = {David L. St-Pierre and Jialin Liu}, pages = {2397--2403}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Differential evolution, Dynamic and uncertain environments}, abstract = { In this paper we compare Differential Evolution (DE), an evolutionary algorithm, to classical bandit algorithms over non-stationary bandit problem. First we define a test case where the variation of the distributions depends on the number of times an option is evaluated rather than over time. This definition allows the possibility to apply these algorithms over a wide range of problems such as black-box portfolio selection. Second we present our own variant of discounted Upper Confidence Bound (UCB) algorithm that outperforms the current state-of-the-art algorithms for non-stationary bandit problem. Third, we introduce a variant of DE and show that, on a selection over a portfolio of solvers for the Cart-Pole problem, our version of DE outperforms the current best UCBs algorithms. }} @InProceedings{Kazimipour:2014:CEC, title = {Effects of Population Initialization on Differential Evolution for Large Scale Optimization}, author = {Borhan Kazimipour and Xiaodong Li and A.K. Qin}, pages = {2404--2411}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems, Differential evolution}, abstract = { This work provides an in-depth investigation of the effects of population initialisation on Differential Evolution (DE) for dealing with large scale Optimisation problems. Firstly, we conduct a statistical parameter sensitive analysis to study the effects of DE's control parameters on its performance of solving large scale problems. This study reveals the optimal parameter configurations which can lead to the statistically superior performance over the CEC-2013 large-scale test problems. Thus identified optimal parameter configurations interestingly favour much larger population sizes while agreeing with the other parameter settings compared to the most commonly employed parameter configuration. Based on one of the identified optimal configurations and the most commonly used configuration, which only differ in the population size, we investigate the influence of various population initialisation techniques on DE's performance. This study indicates that initialisation plays a more crucial role in DE with a smaller population size. However, this observation might be the result of insufficient convergence due to the use of a large population size under the limited computational budget, which deserve more investigations. }} % Session: FrE1-2 Process Mining and Data Mining @InProceedings{vanden-Broucke:2014:CEC, title = {Declarative Process Discovery with Evolutionary Computing}, author = {Seppe vanden Broucke and Jan Vanthienen and Bart Baesens}, pages = {2412--2419}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Process Mining, Genetic algorithms}, abstract = { The field of process mining deals with the extraction of knowledge from event logs. One task within the area of process mining entails the discovery of process models to represent real-life behaviour as observed in day-to-day business activities. A large number of such process discovery algorithms have been proposed during the course of the past decade, among which techniques to mine declarative process models (e.g. Declare and AGNEs Miner) as well as evolutionary based techniques (e.g. Genetic Miner and Process Tree Miner). In this paper, we present the initial results of a newly proposed evolutionary based process discovery algorithm which aims to discover declarative process models, hence combining these two classes (declarative and genetic) of discovery techniques. To do so, we herein use a language bias similar to the one found in AGNEs Miner to allow for the conversion from a set of declarative control-flow based constraints (determining the conditions which have to be satisfied to enable to execution of an activity) to a procedural process model, i.e. a Petri net, though this language bias can be extended to include data-based constraints as well. }} @InProceedings{Burattin:2014:CEC, title = {Control-Flow Discovery from Event Streams}, author = {Andrea Burattin and Alessandro Sperduti and Wil M. P. van der Aalst}, pages = {2420--2427}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Process Mining}, abstract = { Process Mining represents an important research field that connects Business Process Modelling and Data Mining. One of the most prominent task of Process Mining is the discovery of a control-flow starting from event logs. This paper focuses on the important problem of control-flow discovery starting from a stream of event data. We propose to adapt Heuristics Miner, one of the most effective control-flow discovery algorithms, to the treatment of streams of event data. Two adaptations, based on Lossy Counting and Lossy Counting with Budget, as well as a sliding window based version of Heuristics Miner, are proposed and experimentally compared against both artificial and real streams. Experimental results show the effectiveness of control-flow discovery algorithms for streams on artificial and real datasets. }} @InProceedings{Low:2014:CEC, title = {Perturbing Event Logs to Identify Cost Reduction Opportunities: A Genetic Algorithm-Based Approach}, author = {W.Z. Low and J. De Weerdt and M.T. Wynn and A.H.M. ter Hofstede and W.M.P. van der Aalst and S. vanden Broucke}, pages = {2428--2435}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Process Mining}, abstract = { Organisations are constantly seeking new ways to improve operational efficiencies. This research study investigates a novel way to identify potential efficiency gains in business operations by observing how they are carried out in the past and then exploring better ways of executing them by taking into account trade-offs between time, cost and resource utilisation. This paper demonstrates how they can be incorporated in the assessment of alternative process execution scenarios by making use of a cost environment. A genetic algorithm-based approach is proposed to explore and assess alternative process execution scenarios, where the objective function is represented by a comprehensive cost structure that captures different process dimensions. Experiments conducted with different variants of the genetic algorithm evaluate the approach's feasibility. The findings demonstrate that a genetic algorithm-based approach is able to make use of cost reduction as a way to identify improved execution scenarios in terms of reduced case durations and increased resource utilisation. The ultimate aim is to utilise cost-related insights gained from such improved scenarios to put forward recommendations for reducing process-related cost within organisations. }} @InProceedings{Martins:2014:CEC, title = {A Clustering-Based Approach for Exploring Sequences of Compiler Optimizations}, author = {Luiz Martins and Ricardo Nobre and Alexandre Delbem and Eduardo Marques and Joao Cardoso}, pages = {2436--2443}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Classification, clustering and data analysis, Data mining, Intelligent systems applications}, abstract = { In this paper we present a clustering-based selection approach for reducing the number of compilation passes used in search space during the exploration of optimisations aiming at increasing the performance of a given function and/or code fragment. The basic idea is to identify similarities among functions and to use the passes previously explored each time a new function is being compiled. This subset of compiler optimisations is then used by a Design Space Exploration (DSE) process. The identification of similarities is obtained by a data mining method which is applied to a symbolic code representation that translates the main structures of the source code to a sequence of symbols based on transformation rules. Experiments were performed for evaluating the effectiveness of the proposed approach. The selection of compiler Optimisation sequences considering a set of 49 compilation passes and targeting a Xilinx MicroBlaze processor was performed aiming at latency improvements for 41 functions from Texas Instruments benchmarks. The results reveal that the passes selection based on our clustering method achieves a significant gain on execution time over the full search space still achieving important performance speedups. }} @InProceedings{Yoshida:2014:CEC, title = {A Study on Non-Correspondence in Spread between Objective Space and Design Variable Space for Trajectory Designing Optimization Problem}, author = {Toru Yoshida and Tomohiro Yoshikawa}, pages = {2444--2450}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Data mining}, abstract = { Recently, a lot of studies on Multi-Objective Genetic Algorithm (MOGA), in which Genetic Algorithm is applied to Multi-objective Optimisation Problems (MOPs), have been reported actively.MOGA has been also applied to engineering design fields, then it is important not only to obtain Pareto solutions having high performance but also to analyse the obtained Pareto solutions and extract the knowledge in the designing problem.In order to analyse Pareto solutions obtained by MOGA, it is required to consider both the objective space and the design variable space. In this paper, we define"Non-Correspondence in Spread" between the objective space and the design variable space.We also try to extract Non-Correspondence area in Spread with the index defined in this paper.This paper applies the proposed method to the trajectory designing Optimisation problem and extracts Non-Correspondence area in Spread in the acquired Pareto solutions. }} @InProceedings{Agapitos:2014:CEC, title = {Ensemble {Bayesian} Model Averaging in Genetic Programming}, author = {Alexandros Agapitos and Michael O'Neill and Anthony Brabazon}, pages = {2451--2458}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Data mining, Classification, clustering and data analysis}, abstract = { This paper considers the general problem of function estimation via Genetic Programming (GP). Data analysts typically select a model from a population of models, and then proceed as if the selected model had generated the data. This approach ignores the uncertainty in model selection, leading to over-confident inferences and lack of generalisation. We adopt a coherent method for accounting for this uncertainty through a weighted averaging of all models competing in a population of GP. It is a principled statistical method for post-processing a population of programs into an ensemble, which is based on Bayesian Model Averaging (BMA). Under two different formulations of BMA, the predictive probability density function (PDF) of a response variable is a weighted average of PDFs centred around the individual predictions of component models that take the form of either standalone programs or ensembles of programs. The weights are equal to the posterior probabilities of the models generating the predictions, and reflect the models' skill on the training dataset. The method was applied to a number of synthetic symbolic regression problems, and results demonstrate that it generalises better than standard methods for model selection, as well as methods for ensemble construction in GP. }} % Session: FrE1-3 Estimation of Distribution Algorithms and Machine Learning @InProceedings{Ceberio:2014:CEC, title = {Extending Distance-Based Ranking Models in Estimation of Distribution Algorithms}, author = {Josu Ceberio and Ekhine Irurozki and Alexander Mendiburu and Jose A. Lozano}, pages = {2459--2466}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Estimation of distribution algorithms, Discrete and combinatorial Optimisation}, abstract = { Recently, probability models on rankings have been proposed in the field of estimation of distribution algorithms in order to solve permutation-based combinatorial optimisation problems. Particularly, distance-based ranking models, such as Mallows and Generalised Mallows under the Kendall's-tau distance, have demonstrated their validity when solving this type of problems. Nevertheless, there are still many trends that deserve further study. In this paper, we extend the use of distance-based ranking models in the framework of EDAs by introducing new distance metrics such as Cayley and Ulam. In order to analyse the performance of the Mallows and Generalised Mallows EDAs under the Kendall, Cayley and Ulam distances, we run them on a benchmark of 120 instances from four well known permutation problems. The conducted experiments showed that there is not just one metric that performs the best in all the problems. However, the statistical test pointed out that Mallows-Ulam EDA is the most stable algorithm among the studied proposals. }} @InProceedings{Wang:2014:CECi, title = {Quantum-Inspired Evolutionary Algorithm with Linkage Learning}, author = {Bo Wang and Hua Xu and Yuan Yuan}, pages = {2467--2474}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Estimation of distribution algorithms, Discrete and combinatorial Optimisation}, abstract = { The quantum-inspired evolutionary algorithm (QEA) uses several quantum computing principles to optimise problems on a classical computer. QEA possesses a number of quantum individuals, which are all probability vectors. They work well for linear problems but fail on problems with strong interactions among variables. Moreover, many Optimisation problems have multiple global optima. And because of the genetic drift, these problems are difficult for evolutionary algorithms to find all global optima. Local and global migration that QEA uses to synchronise different individuals prevent QEA from finding multiple optima. To overcome these difficulties, we proposed a quantum-inspired evolutionary algorithm with linkage learning (QEALL). QEALL uses a modified concept-guide operator based on low order statistics to learn linkage. We also replaced the migration procedure by a niching technology to prevent genetic drift, accordingly to find all global optima and to expedite convergence speed. The performance of QEALL was tested on a number of benchmarks including both unimodal and multimodal problems. Empirical evaluation suggests that the proposed algorithm is effective and efficient. }} @InProceedings{Wang:2014:CECj, title = {Investigation on Efficiency of Optimal Mixing on Various Linkage Sets}, author = {Shih-Ming Wang and Yu-Fan Tung and Tian-Li Yu}, pages = {2475--2482}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Estimation of distribution algorithms, Genetic algorithms}, abstract = { The optimal mixing operator (OM) uses linkage sets (LSs) to exchange the information of variables between a pair of solutions, and the result of such exchange is adopted only if the exchange leads to improvement of the solution quality. The performance of OM highly depends on the LS it uses. This paper demonstrates that previously proposed LS, the linkage tree model (LT), does not yield the optimal performance. To measure the efficiency of OM on different LSs, the cost-performance (CP) index is defined. Both our CP index and experiments indicate (1) that for fully separable problems, the most suitable LS is the marginal product model (MP), and (2) that for separable problems with overlap, LT is more suitable than MP, and (3) that properly pruned LT leads to higher efficiency and yields a better performance, and (4) that the LS that properly reflects the problem structure yields the best performance on both fully separable problems and problems with overlap. }} @InProceedings{Liao:2014:CEC, title = {A Locally Weighted Metamodel for Pre-Selection in Evolutionary Optimization}, author = {Qiuxiao Liao and Aimin Zhou and Guixu Zhang}, pages = {2483--2490}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Algorithms with Statistical and Machine Learning Techniques, Meta-modelling and surrogate models}, abstract = { The evolutionary algorithms are usually criticised for their slow convergence. To address this weakness, a variety of strategies have been proposed. Among them, the meta-model or surrogate based approaches are promising since they replace the original Optimisation objective by a metamodel. However, the metamodel building itself is expensive and therefore the metamodel based evolutionary algorithms are commonly applied to expensive Optimisation. In this paper, we propose an alternative metamodel, named locally weighted metamodel (LWM), for the pre-selection in evolutionary Optimisation. The basic idea is to estimate the objective values of candidate offspring solutions for an individual, and choose the most promising one as the offspring solution. Instead of building a global model as many other algorithms do, a LWM is built for each candidate offspring solution in our approach. The LWM based pre-selection is implemented in a multi-operator based evolutionary algorithm, and applied to a set of test instances with different characteristics. Experimental results show that the proposed approach is promising. }} @InProceedings{Su:2014:CEC, title = {Use Model Building on Discretization Algorithms for Discrete {EDAs} to Work on Real-Valued Problems}, author = {Yi-En Su and Tian-Li Yu}, pages = {2491--2498}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Evolution strategies}, abstract = { Discretisation algorithms have been combined with discrete estimation of distribution algorithms (EDAs) to work on real-valued problems. Existing discretisation algorithms, such as the fixed-height histogram (FHH) and the split-on-demand (SoD), use merely densities of selected chromosomes to build next generation population, and therefore have limited exploration. This paper adds the concept of model building to FHH and SoD to solve these problems. The model uses a variety of information from selected chromosomes to improve the abilities of FHH and SoD to identify promising regions for future exploration. Specifically, a model of expected values of selected chromosomes is combined with FHH and SoD to form expected-value FHH and expected-value SoD. The expected-value-discretisation algorithms outperform their original versions on an exploration test function as well as the 25 benchmark functions used in the SoD paper. This paper also introduces a model of differential-expected-value of selected chromosomes. The differential-expected-value FHH and differential-expected-value SoD outperform their expected-value versions when tested on the exploration test function and the 25 benchmark functions. }} @InProceedings{Kattan:2014:CEC, title = {Transformation of Input Space Using Statistical Moments: {EA}-Based Approach}, author = {Ahmed Kattan and Michael Kampouridis and Yew-Soon Ong and Khalid Mehamdi}, pages = {2499--2506}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Genetic programming}, abstract = { Reliable regression models in the field of Machine Learning (ML) revolve around the fundamental property of generalisation. This ensures that the induced model is a concise approximation of a data-generating process and performs correctly when presented with data that have not been utilised during the learning process. Normally, the regression model is presented with n samples from an input space; that is composed of observational data of the form (xi, y(xi)), i = 1...n where each xi denotes a k dimensional input vector of design variables and y is the response. When k n, high variance and over-fitting become a major concern. In this paper we propose a novel approach to mitigate this problem by transforming the input vectors into new smaller vectors (called Z set) using only a set of simple statistical moments. Genetic Algorithm (GA) has been used to evolve a transformation procedure. It is used to optimise an optimal sequence of statistical moments and their input parameters. We used Linear Regression (LR) as an example to quantify the quality of the evolved transformation procedure. Empirical evidences, collected from benchmark functions and real-world problems, demonstrate that the proposed transformation approach is able to dramatically improve LR generalisation and make it outperform other state of the art regression models such as Genetic Programming, Kriging, and Radial Basis Functions Networks. In addition, we present an analysis to shed light on the most important statistical moments that are useful for the transformation process. }} % Session: FrE1-4 Evolutionary Computation Theory and Parameter Optimisation @InProceedings{Malan:2014:CEC, title = {A Progressive Random Walk Algorithm for Sampling Continuous Fitness Landscapes}, author = {Katherine Malan and Andries Engelbrecht}, pages = {2507--2514}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary computation theory}, abstract = { A number of fitness landscape analysis approaches are based on random walks through discrete search spaces. Applying these approaches to real-encoded problems requires the notion of a random walk in continuous space. This paper proposes a progressive random walk algorithm and the use of multiple walks to sample neighbourhood structure in continuous multi-dimensional spaces. It is shown that better coverage of a search space is provided by progressive random walks than simple unbiased random walks. }} @InProceedings{Alanazi:2014:CEC, title = {Runtime Analysis of Selection Hyper-Heuristics with Classical Learning Mechanisms}, author = {Fawaz Alanazi and Per Kristian Lehre}, pages = {2515--2523}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Evolutionary computation theory}, abstract = { The term selection hyper-heuristics refers to a randomised search technique used to solve computational problems by choosing and executing heuristics from a set of pre-defined low-level heuristic components. Selection hyper-heuristics have been successfully employed in many problem domains. Nevertheless, a theoretical foundation of these heuristics is largely missing. Gaining insight into the behaviour of selection hyper-heuristics is challenging due to the complexity and random design of these heuristics. This paper is one of the initial studies to analyse rigorously the runtime of selection hyper-heuristics with a number of the most commonly used learning mechanisms; namely, simple random, random gradient, greedy, and permutation. We derive the runtime of selection hyperheuristic with these learning mechanisms not only on a classical example problem, but also on a general model of fitness landscapes. This in turn helps in understanding the behaviour of hyper-heuristics. Our results show that all the considered selections hyper-heuristics have roughly the same performance. This suggests that the learning mechanisms do not necessarily improve the performance of hyper-heuristics. A new learning mechanism that improves the performance of hyper-heuristic on our example problem is presented. }} @InProceedings{Cleghorn:2014:CEC, title = {Particle Swarm Convergence: An Empirical Investigation}, author = {Christopher Cleghorn and Andries Engelbrecht}, pages = {2524--2530}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Convergence, scalability and complexity analysis, Particle swarm Optimisation}, abstract = { This paper performs a thorough empirical investigation of the conditions placed on particle swarm Optimisation control parameters to ensure convergent behaviour. At present there exists a large number of theoretically derived parameter regions that will ensure particle convergence, however, selecting which region to use in practice is not obvious. The empirical study is carried out over a region slightly larger than that needed to contain all the relevant theoretically derived regions. It was found that there is a very strong correlation between one of the theoretically derived regions and the empirical evidence. It was also found that parameters near the edge of the theoretically derived region converge at a very slow rate, after an initial population explosion. Particle convergence is so slow, that in practice, the edge parameter settings should not really be considered useful as convergent parameter settings. }} @InProceedings{Ma:2014:CECc, title = {Phase Transition Particle Swarm Optimization}, author = {Ji Ma and Junqi Zhang and Wei Wang and Jing Yao}, pages = {2531--2538}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm Optimisation (PSO), Self-adaptation in evolutionary computation, Numerical Optimisation}, abstract = { In nature, a phase transition is the transformation of a thermodynamic system from one phase to another. Different phases of a thermodynamic system have distinctive physical properties. Inspired by this natural phenomenon, this paper presents a Particle Swarm Optimisation (PSO) based on the Phase Transitions model which consists of solid, liquid and gas phases. Each phase represents a distinctive behaviour of the swarm. Transitions of condensation, solidification and deposition can enhance the exploitation capability of the swarm. While the transitions of fusion, vaporisation and sublimation from the other direction improve the exploration capability of the swarm. The proposed model directs the swarm to transform among phases dynamically and automatically according to the evolutional states to balance between exploration and exploitation adaptively. Especially, it uses a new modified PSO algorithm called Simple Fast Particle Swarm Optimisation (SFPSO) in the solid phase, which modifies the original PSO by adding new parameters simply to make the algorithm convergence more quickly. The proposed algorithm is validated by extensive simulations on the 28 real-parameter Optimisation benchmark functions from CEC 2013 compared with other three representative variants of PSO. }} @InProceedings{Zhang:2014:CECi, title = {Fitness Level Based Adaptive Operator Selection for Cutting Stock Problems with Contiguity}, author = {Kai Zhang and Thomas Weise and Jinlong Li}, pages = {2539--2546}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Discrete and combinatorial Optimisation, Evolution strategies, Self-adaptation in evolutionary computation}, abstract = { For most Optimisation problem and solution representation, multiple different possible search operators exist. In this article, we propose the Fitness Level based Adaptive Operator Selection (FLAOS), a self-adaptation approach that automatically selects the right operator depending on the progress of the search. In FLAOS, the objective values of the solutions discovered during the Optimisation process are divided into intervals, the fitness levels. For each fitness level, a corresponding probability distribution is maintained which defines which operators are to be used and how often to generate the children. An evolutionary algorithm with FLAOS is proposed to solve one-dimensional cutting stock problems (CSPs) with contiguity. The solution of such a problem should minimise both the trim loss and the number of partially finished items. Experimental studies have been carried out to test the effectiveness of the FLAOS. The solutions found by FLAOS are better than or comparable to those solutions found by previous methods. }} @InProceedings{Klazar:2014:CEC, title = {Parameter Optimization by Means of Statistical Quality Guides in {F-Race}}, author = {Ronald Klazar and Andries Engelbrecht}, pages = {2547--2552}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical Optimisation}, abstract = { F-Race and its variant, Iterated F-Race, is an automated procedure for sampling and evaluating potential values of parameters for algorithms. The procedure is controlled by means of a computational budget that limits the number of evaluations that may be conducted, thus forcing the determination of the best possible configuration to be made within a limited time. When time is not severely constrained, the a priori choice of a computational budget becomes unjustifiable because the relationship between the computational budget and the quality of the Optimisation of a black box subject is not obvious. This paper proposes an extension to F-Race in the form of a heuristic method for reasonably terminating the Optimisation procedure. }} % Session: FrE1-5 Multimodal Optimisation and Population Initialisation @InProceedings{Zhang:2014:CECj, title = {A Globally Diversified Island Model {PGA} for Multimodal Optimization}, author = {Lifeng Zhang and Rong He}, pages = {2553--2561}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Niching Methods for Multimodal Optimisation, Parallel and distributed algorithms}, abstract = { Multimodal Optimisation aims to find multiple global and local optima as opposed to only the best optimum. Parallel genetic algorithms (PGAs) provide a natural advantage for dealing with this issue, since they are multi-population based searching methodologies. For single population based evolutionary algorithms, a number of niching and multimodal Optimisation techniques have been proposed and successfully applied to cope with this problem. However, these approaches are definitely not applicable for PGAs, since due to communicational and computational costs it is very always impossible to obtain and compute global information of all the sub-populations during massive parallel evolution procedure. In this study, a new island model PGA, called local competition model (LCM), is developed to cope with this issue. The new method only uses local information received from a few neighbouring subpopulations to reach a global diversification in which all the subpopulations are automatically allocated to different areas of searching space so that they can converge to multiple optima including both global optima and local optima. Finally, experimental studies on both real number Optimisation and combinatorial Optimisation are implemented to illustrate the performance of the new PGA model. }} @InProceedings{Pereira:2014:CEC, title = {A Topological Niching Covariance Matrix Adaptation for Multimodal Optimization}, author = {Marcio Pereira and Mauro Roisenberg and Guenther Neto}, pages = {2562--2569}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Niching Methods for Multimodal Optimisation, Evolution strategies, Multi-objective evolutionary algorithms}, abstract = { Multimodal Optimisation attempts to find multiple global and local optima of a function. Finding a set of optimal solutions is particularly important for practical problems. However, this kind of problem requires Optimisation techniques that demand a high computational cost and a large amount of parameters to be adjusted. These difficulties increase in high-dimensional space problems. In this work, we propose a niching method based on recent developments in the basins (optimal locations) identification to reduce costs and perform better in high-dimensional spaces. Using Nearest-Better Clustering (NBC) and Hill-Valley (or Detect Multimodal) methods, an exploratory initialisation routine is employed to identify basins on functions with different levels of complexity. To maintain diversity over the generations, we define a bi-objective function, which is composed by the original fitness function and the distance to the nearest better neighbour, assisted by a reinitialisation scheme. The proposed method is implemented using Evolutionary Strategy (ES) known as Covariance Matrix Adaptation (CMA). Unlike recent multimodal approaches using CMA-ES, we use its step size to control the influence of niche, thus avoiding extra efforts in parametrisation. We apply a benchmark of 20 test functions, specially designed for multimodal Optimisation evaluation, and compare the performance with a state-of-the-art method. Finally we discuss the results and show that the proposed approach can reach better and stable results even in high-dimensional spaces. }} @InProceedings{Vafaee:2014:CEC, title = {Balancing the Exploration and Exploitation in an Adaptive Diversity Guided Genetic Algorithm}, author = {Fatemeh Vafaee and Gyorgy Turan and Peter Nelson and Tanya Berger-Wolf}, pages = {2570--2577}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms}, abstract = { Exploration and exploitation are the two cornerstones which characterise Evolutionary Algorithms (EAs) capabilities. Maintaining the reciprocal balance of the explorative and exploitative power is the key to the success of EA applications. Accordingly, this work is concerned with proposing a diversity guided genetic algorithm with a new mutation scheme that is capable of exploring the unseen regions of the search space, as well as exploiting the already-found promising elements. The proposed mutation operator specifies different mutation rates for different sites of an encoded solution. These site-specific rates are carefully derived based on the underlying pattern of highly-fit solutions, adjusted to every single individual, and adapted throughout the evolution to retain a good ratio between exploration and exploitation. Furthermore, in order to more directly monitor the exploration vs. exploitation balance, the proposed method is augmented with a diversity control process assuring that the search process does not lose the required balance between the two forces. }} @InProceedings{Peng:2014:CECa, title = {Compensate Information from Multimodal Dynamic Landscapes: An Anti-Pathology Cooperative Coevolutionary Algorithm}, author = {Xingguang Peng and Xiaokang Lei and Kun Liu}, pages = {2578--2584}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Collaborative Learning and Optimisation, Coevolution and collective behaviour, Dynamic and uncertain environments}, abstract = { Cooperative coevolutionary algorithms (CCEAs) divides a problem into several components and optimises them independently. Some coevolutionary information will be lost due to the search space separation. This may lead some algorithmic pathologies, such as relative over generalisation. In addition, according to the interactive nature of the CCEA, the coevolutionary landscapes are dynamic. In this paper, a multi-population strategy is proposed to simultaneously search local or global optima in each dynamic landscape and provide them to the other components. Besides, a grid-based archive scheme is proposed to archive these historic collaborators for reasonable fitness evaluation. Two benchmark problems were used to test and compare the proposed algorithm to three classical CCEAs. Experimental results show that the proposed algorithm effectively counteract relative over generalisation pathology and significantly improve the rate of converging to global optimum. }} @InProceedings{Kazimipour:2014:CECa, title = {A Review of Population Initialization Techniques for Evolutionary Algorithms}, author = {Borhan Kazimipour and Xiaodong Li and A.K. Qin}, pages = {2585--2592}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems, Representation and operators}, abstract = { Although various population initialisation techniques have been employed in evolutionary algorithms (EAs), there lacks a comprehensive survey on this research topic. To fill this gap and attract more attentions from EA researchers to this crucial yet less explored area, we conduct a systematic review of the existing population initialisation techniques. Specifically, we categorise initialisation techniques from three exclusive perspectives, i.e., randomness, compositionality and generality. Characteristics of the techniques belonging to each category are carefully analysed to further lead to several sub-categories. We also discuss several open issues related to this research topic, which demands further in-depth investigations. }} @InProceedings{Fieldsend:2014:CEC, title = {Running Up Those Hills: Multi-Modal Search with the Niching Migratory Multi-Swarm Optimiser}, author = {Jonathan Fieldsend}, pages = {2593--2600}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Niching Methods for Multimodal Optimisation}, abstract = { We present a new multi-modal evolutionary optimiser, the niching migratory multi-swarm optimiser (NMMSO), which dynamically manages many particle swarms. These sub-swarms are concerned with optimising separate local modes, and employ measures to allow swarm elements to migrate away from their parent swarm if they are identified as being in the vicinity of a separate peak, and to merge swarms together if they are identified as being concerned with the same peak. We employ coarse peak identification to facilitate the mode identification required. Swarm members are not constrained to particular subregions of the parameter space, however members are initialised in the vicinity of a swarm's local mode estimate. NMMSO is shown to cope with a range of problem types, and to produce results competitive with the state-of-the-art on the CEC 2013 multi-modal optimisation competition test problems, providing new benchmark results in the field. }} % Session: FrE2-1 Multi-Objective Evolutionary Algorithms III @InProceedings{Zhu:2014:CECb, title = {Multi-Scenario Optimization Using Multi-Criterion Methods: A Case Study on {Byzantine} Agreement Problem}, author = {Ling Zhu and Kalyanmoy Deb and Sandeep Kulkarni}, pages = {2601--2608}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Evolutionary simulation-based Optimisation}, abstract = { In this paper, we address solution methodologies of an Optimisation problem under multiple scenarios. Often in practice, a problem needs to be considered for different scenarios, such as evaluating for different loading conditions, different blocks of data, multi-stage operations, etc. After reviewing various single-objective aggregate methods for handling objectives and constraints under multiple scenarios, we then suggest a multi-objective Optimisation approach for solving multi-scenario Optimisation problems. On a Byzantine agreement problem, we demonstrate the usefulness of the proposed multi-objective approach and explain the reasons for their superior behaviour. The suggested procedure is generic and now awaits further applications to more challenging problems from engineering and computational fields. }} @InProceedings{Smith:2014:CEC, title = {Multi-Objective Evolutionary Recurrent Neural Network Ensemble for Prediction of Computational Fluid Dynamic Simulations}, author = {Christopher Smith and John Doherty and Yaochu Jin}, pages = {2609--2616}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Surrogate-assisted Global Optimisation Methods for Expensive Engineering Design}, abstract = { Using a surrogate model to evaluate the expensive fitness of candidate solutions in an evolutionary algorithm can significantly reduce the overall computational cost of Optimisation tasks. In this paper we present a recurrent neural network ensemble that is used as a surrogate for the long-term prediction of computational fluid dynamic simulations. A hybrid multi-objective evolutionary algorithm that trains and optimises the structure of the recurrent neural networks is introduced. Selection and combination of individual prediction models in the Pareto set of solutions is used to create the ensemble of predictors. Five selection methods are tested on six data sets and the accuracy of the ensembles is compared to the converged computational fluid dynamic data, as well as to the delta change between two flow conditions. Intermediate computational fluid dynamic data is used for training and the method presented can produce accurate and stable results using a third of the intermediate data needed for convergence. }} @InProceedings{Wesolkowski:2014:CEC, title = {{TraDE}: Training Device Selection Via Multi-Objective Optimization}, author = {Slawomir Wesolkowski and Nevena Francetic and Stuart Grant}, pages = {2617--2624}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristic Methods for Multi-Component Optimisation Problems, Multi-objective evolutionary algorithms,NSGA-II, Real-world applications}, abstract = { Training planning is a recurring military problem. Since training programs can use multiple training devices with varying costs and training capabilities, selecting the types of devices required is a complex trade-off problem. Furthermore, the placement of these devices is critical due to the time and costs involved in travelling to and from the location of a training device. In this paper, we introduce a device bin-packing-and-location-based model, Training Device Estimation (TraDE), to study the computation of heterogeneous device mixes including the location of each device with respect to numerous objectives including various costs and training time. We apply the multi-objective Non-dominating Sorting Genetic Algorithm II to the TraDE model on a population represented by two-dimensional chromosomes. Finally, we also present a new mutation type to handle the nonlinearity inherent in a dual Optimisation problem which includes scheduling and location Optimisation. We clearly show that the new mutation operator produces superior results to the standard mutation operator. }} @InProceedings{Abdul:2014:CEC, title = {Multi-view Clustering of Web Documents Using Multi-Objective Genetic Algorithm}, author = {Wahid Abdul and Gao Xiaoying and Andreae Peter}, pages = {2625--2632}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation for Grouping and Graph-based Clustering}, abstract = { Clustering ensembles are a common approach to clustering problem, which combine a collection of clustering into a superior solution. The key issues are how to generate different candidate solutions and how to combine them. Common approach for generating candidate clustering solutions ignores the multiple representations of the data (i.e., multiple views) and the standard approach of simply selecting the best solution from candidate clustering solutions ignores the fact that there may be a set of clusters from different candidate clustering solutions which can form a better clustering solution. This paper presents a new clustering method that exploits multiple views to generate different clustering solutions and then selects a combination of clusters to form a final clustering solution. Our method is based on Non dominated Sorting Genetic Algorithm (NSGA-II), which is a multi-objective Optimisation approach. Our new method is compared with five existing algorithms on three data sets that have increasing difficulty. The results show that our method significantly outperforms other methods. }} @InProceedings{Masuda:2014:CEC, title = {Visual Examination of the Behavior of {EMO} Algorithms for Many-Objective Optimization with Many Decision Variables}, author = {Hiroyuki Masuda and Yusuke Nojima and Hisao Ishibuchi}, pages = {2633--2640}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making, Multi-objective evolutionary algorithms}, abstract = { Various evolutionary multiobjective Optimisation (EMO) algorithms have been proposed in the literature. They have different search mechanisms for increasing the diversity of solutions and improving the convergence to the Pareto front. As a result, each algorithm has different characteristics in its search behaviour. Multiobjective search behaviour can be visually shown in an objective space for a test problem with two or three objectives. However, such a visual examination is difficult in a high-dimensional objective space for many-objective problems. The use of distance minimisation problems has been proposed to examine many-objective search behaviour in a two-dimensional decision space. This idea has an inherent limitation: the number of decision variables should be two. In our former study, we formulated a four-objective distance minimisation problem with 10, 100, and 1000 decision variables. In this paper, we generalise our former study to many-objective problems with an arbitrarily number of objectives and decision variables by proposing an idea of specifying reference points on a plane in a high-dimensional decision space. As test problems for computational experiments, we generate six-objective and eight-objective problems with 10, 100, and 1000 decision variables. Our experimental results on those test problems show that the number of decision variables has large effects on multiobjective search in comparison with the choice of an EMO algorithm and the number of objectives. }} @InProceedings{Hu:2014:CECc, title = {Sensitivity Analysis of Parallel Cell Coordinate System in Many-Objective Particle Swarm Optimization}, author = {Wang Hu and Gary Yen and Xin Zhang}, pages = {2641--2648}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm Optimisation (PSO), Multi-objective evolutionary algorithms, Large-scale problems}, abstract = { Parallel Cell Coordinate System (PCCS) was proposed to evaluate the individual fitness in an archive and access the population progress in the evolutionary environment. In a Many-objective Optimisation Problem (MaOP), it is much harder to trade off the convergence and diversity than in a Multiobjective Optimisation Problem. To more effectively tackle the MaOPs, the PCCS and the aggregation based approach are integrated into a Many-objective Optimisation Particle Swarm Optimisation (MaOPSO). In this paper, the sensitivity of PCCS is examined with respect to the number of objectives and the maximum size of an archive. The experimental results indicate that the MaOPSO performs better than MOEA/D in terms of IGD and HV metrics on the WFG test suit, and PCCS is not sensitive to the number of objectives and the maximum size of an archive. }} % Session: FrE2-2 Numerical Optimisation @InProceedings{Maia:2014:CEC, title = {Real-Parameter Optimization with {OptBees}}, author = {Renato Maia and Leandro de Castro and Walmir Caminhas}, pages = {2649--2655}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Numerical Optimisation}, abstract = { This paper reports how OptBees, an algorithm inspired by the collective decision-making of bee colonies, performed in the test bed developed for the Special Session and Competition on Real-Parameter Single Objective (Expensive) Optimisation at CEC-2014. The test bed includes 30 scalable functions, many of which are both non-separable and highly multi-modal. Results include OptBees' performance on the 10, 30, 50 and 100-dimensional versions of each function. }} @InProceedings{Shan:2014:CEC, title = {A {Levy} Flight-Based Hybrid Artificial Bee Colony Algorithm for Solving Numerical Optimization Problems}, author = {Hai Shan and Toshiyuki Yasuda and Kazuhiro Ohkura}, pages = {2656--2663}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical Optimisation, Heuristics, metaheuristics and hyper-heuristics}, abstract = { An artificial bee colony (ABC) algorithm is one of numerous swarm intelligence algorithms that employs the foraging behaviour of honeybee colonies. To improve the convergence performance and search speed of finding the best solution using this approach, we propose a levy flight-based hybrid ABC algorithm in this paper. To evaluate the performance of the standard and proposed ABC algorithms, we implemented numerical Optimisation problems based on the IEEE Congress on Evolutionary Computation 2013 test suite. The proposed ABC algorithm demonstrated competitive performance on these Optimisation problems as compared to standard ABC, differential evolution, and particle swarm Optimisation algorithms with dimension sizes of 10, 30, and 50, respectively. }} @InProceedings{Ding:2014:CECb, title = {Comparison of Random Number Generators in Particle Swarm Opimization Algorithm}, author = {Ke Ding and Ying Tan}, pages = {2664--2671}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm Optimisation (PSO), Heuristics, metaheuristics and hyper-heuristics, Numerical Optimisation}, abstract = { Intelligent Optimisation algorithms are very effective to tackle complex problems that would be difficult or impossible to solve exactly. A key component within these algorithms is the random number generators (RNGs) which provide random numbers to drive the stochastic search process. Much effort is devoted to develop efficient RNGs with good statistical properties, and many highly optimised libraries are ready to use for generating random numbers fast on both CPUs and other hardware platforms such as GPUs. However, few study is focused on how different RNGs can effect the performance of specific intelligent Optimisation algorithms. In this paper, we empirically compared 13 widely used RNGs with uniform distribution based on both CPUs and GPUs, with respect to algorithm efficiency as well as their impact on Particle Swarm Optimisation (PSO). Two strategies were adopted to conduct comparison among multiple RNGs for multiple objectives. The experiments were conducted on well-known benchmark functions of diverse landscapes, and were run on the GPU for the purpose of accelerating. The results show that RNGs have very different efficiencies in terms of speed, and GPU-based RNGs can be much faster than their CPU-based counterparts if properly used. However, no statistically significant disparity in solution quality was observed. Thus it is reasonable to use more efficient RNGs such as Mersenne Twister. The framework proposed in this work can be easily extended to compare the impact of non-uniformly distributed RNGs on more other intelligent Optimisation algorithms. }} @InProceedings{Chen:2014:CECd, title = {A Evolutionary Algorithm Based on Covariance Matrix Leaning and Searching Preference for Solving {CEC 2014} Benchmark Problems}, author = {Lei Chen and Hai-Lin Liu and Zhe Zheng and Shengli Xie}, pages = {2672--2677}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation, Genetic algorithms, Numerical Optimisation}, abstract = { In this paper, we propose a single objective Optimisation evolutionary algorithm (EA) based on Covariance Matrix Learning and Searching Preference (CMLSP) and design a switching method which is used to combine CMLSP and Covariance Matrix Adaptation Evolution Strategy (CMAES). Then we investigate the performance of the switch method on a set of 30 noiseless Optimisation problems designed for the special session on real-parameter Optimisation of CEC 2014. The basic idea of the proposed CMLSP is that it is more likely to find a better individual around a good individual. That is to say, the better an individual is, the more resources should be invested to search the region around the individual. To achieve it, we discard the traditional crossover and mutation and design a novel method based on the covariance matrix leaning to generate high quality solutions. The best individual found so far is used as the mean of a Gaussian distribution and the covariance of the best individuals in the population are used as the evaluation of its covariance matrix and we sample the next generation individual from the Gaussian distribution other than using crossover and mutation. In the process of generating new individuals, the best individual is changed if ever a better one is found. This search strategy emphasises the region around the best individual so that a faster convergence can be achieved. The use of switch method is to make best use of the proposed CMLSP and existing CMAES. At last, we report the results. }} @InProceedings{Leite:2014:CEC, title = {Optimization of Power Flow with Energy Storage Using Genetic Algorithms}, author = {Vitor Leite and Carlos Silva and Joao Claro and Joao M. C. Sousa}, pages = {2678--2684}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Engineering applications, Genetic algorithms}, abstract = { This paper applies genetic algorithms to optimise the operation of a transmission network with energy storage capabilities, to optimise its costs, which include both generation and storage costs, for cases when the data inherent to the system is assumed to be perfectly known. The problem is formulated through the DC optimal power flow equations, including losses across the transmission lines, therefore allowing solutions regarding the network generation costs to be obtained, with and without storage. In this way, the financial impact inherent to the usage of energy storage can be derived. Since we are dealing with a large combinatorial problem, the search throughout the solution space was done by means of the Genetic Algorithms. The solutions consist of the storage device's charging or discharging rate at which it must be operating during each sub-interval considered for the simulations. The results delivered by the GA have proved the profitability of including energy storage capabilities in the transmission network of Sao Miguel (Portugal) and the usefulness of such algorithm in a real world application. }} @InProceedings{Yang:2014:CECc, title = {A New Self-Learning {TLBO} Algorithm for {RBF} Neural Modelling of Batteries in Electric Vehicles}, author = {Zhile Yang and Kang Li and Aoife Foley and Cheng Zhang}, pages = {2685--2691}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolved neural networks, Engineering applications, Heuristics, metaheuristics and hyper-heuristics}, abstract = { One of the main purposes of building a battery model is for monitoring and control during battery charging/discharging as well as for estimating key factors of batteries such as the state of charge for electric vehicles. However, the model based on the electrochemical reactions within the batteries is highly complex and difficult to compute using conventional approaches. Radial basis function (RBF) neural networks have been widely used to model complex systems for estimation and control purpose, while the Optimisation of both the linear and non-linear parameters in the RBF model remains a key issue. A recently proposed meta-heuristic algorithm named Teaching-Learning-Based Optimisation (TLBO) is free of presetting algorithm parameters and performs well in non-linear Optimisation. In this paper, a novel self-learning TLBO based RBF model is proposed for modelling electric vehicle batteries using RBF neural networks. The modelling approach has been applied to two battery testing data sets and compared with some other RBF based battery models, the training and validation results confirm the efficacy of the proposed method. }} % Session: FrE2-3 Coevolution and Collective Behaviour @InProceedings{Richter:2014:CEC, title = {Codynamic Fitness Landscapes of Coevolutionary Minimal Substrates}, author = {Hendrik Richter}, pages = {2692--2699}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Coevolutionary systems, Coevolution and collective behaviour, Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { Coevolutionary minimal substrates are simple and abstract models that allow studying the relationships and co-dynamics between objective and subjective fitness. Using these models an approach is presented for defining and analysing fitness landscapes of coevolutionary problems. We devise similarity measures of codynamic fitness landscapes and experimentally study minimal substrates of test--based and compositional problems for both cooperative and competitive interaction. }} @InProceedings{Dick:2014:CEC, title = {Model Representation and Cooperative Coevolution for Finite-State Machine Evolution}, author = {Grant Dick and Xin Yao}, pages = {2700--2707}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, FSM, Evolutionary programming, Coevolutionary systems, Coevolution and collective behaviour}, abstract = { The use and search of finite-state machine (FSM) representations has a long history in evolutionary computation. The flexibility of Mealy-style and Moore-style FSMs is traded against the large number of parameters required to encode machines with many states and/or large output alphabets. Recent work using Mealy FSMs on the Tartarus problem has shown good performance of the resulting machines, but the evolutionary search is slower than for other representations. The aim of this paper is two-fold: first, a comparison between Mealy and Moore representations is considered on two problems, and then the impact of cooperative coevolution on FSM evolutionary search is examined. The results suggest that the search space of Moore-style FSMs may be easier to explore through evolutionary search than the search space of an equivalent-sized Mealy FSM representation. The results presented also suggest that the tested cooperative coevolutionary algorithms struggle to appropriately manage the non-separability present in FSMs, indicating that new approaches to cooperative coevolution may be needed to explore FSMs and similar graphical structures. }} @InProceedings{Wu:2014:CECe, title = {Evolutionary Path Planning of a Data Mule in Wireless Sensor Network by Using Shortcuts}, author = {Shao-You Wu and Jing-Sin Liu}, pages = {2708--2715}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Evolutionary simulation-based Optimisation, Coevolution and collective behaviour}, abstract = { Data collection problem of generating a path for a data mule (single or multiple mobile robots) to collect data from wireless sensor network (WSN) is usually a NP-hard problem. Thus, we formulate it as a Travelling Salesman Problem with Neighbourhoods (TSPN) to obtain the possibly short path. TSPN is composed of determinations of the order of visiting sites and their precise locations. By taking advantage of the overlap of neighbourhoods, we proposed a clustering-based genetic algorithm (CBGA) with an innovative way for initial population generation, called Balanced Standard Deviation Algorithm (BSDA). Then, effective shortcut schemes named Look-Ahead Locating Algorithm (LLA) and Advanced-LLA are applied on the TSPN route. By LLA, a smoother route is generated and the data mule can move while ignoring about 39\% clusters. Extensive simulations are performed to evaluate the TSPN route in some aspects like LLA hits, LLA improvement, Rotation Degree of Data Mule (RDDM), Max Step and Ruggedness. }} @InProceedings{Karim:2014:CEC, title = {Coevolutionary Genetic Algorithm for Variable Ordering in {CSPs}}, author = {Muhammad Rezaul Karim and Malek Mouhoub}, pages = {2716--2723}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling, Coevolutionary systems, Genetic algorithms}, abstract = { A Constraint Satisfaction Problem (CSP) is a framework used for modelling and solving constrained problems. Tree-search algorithms like backtracking try to construct a solution to a CSP by selecting the variables of the problem one after another. The order in which these algorithm select the variables potentially have significant impact on the search performance. Various heuristics have been proposed for choosing good variable ordering. Many powerful variable ordering heuristics weigh the constraints first and then use the weights for selecting good order of the variables. Constraint weighting are basically employed to identify global bottlenecks in a CSP. In this paper, we propose a new approach for learning weights for the constraints using competitive coevolutionary Genetic Algorithm (GA). Weights learnt by the coevolutionary GA later help to make better choices for the first few variables in a search. In the competitive coevolutionary GA, constraints and candidate solutions for a CSP evolve together through an inverse fitness interaction process. We have conducted experiments on several random, quasi-random and patterned instances to measure the efficiency of the proposed approach. The results and analysis show that the proposed approach is good at learning weights to distinguish the hard constraints for quasi-random instances and forced satisfiable random instances generated with the Model \$RB\$. For other type of instances, RNDI still seems to be the best approach as our experiments show. }} @InProceedings{Menendez:2014:CEC, title = {A Co-Evolutionary Multi-Objective Approach for a K-Adaptive Graph-Based Clustering Algorithm}, author = {Hector D. Menendez and David F. Barrero and David Camacho}, pages = {2724--2731}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Multi-objective evolutionary algorithms, Data mining}, abstract = { Clustering is a field of Data Mining that deals with the problem of extract knowledge from data blindly. Basically, clustering identifies similar data in a dataset and groups them in sets named clusters. The high number of clustering practical applications has made it a fertile research topic with several approaches. One recent method that is gaining popularity in the research community is Spectral Clustering (SC). It is a clustering method that builds a similarity graph and applies spectral analysis to preserve the data continuity in the cluster. This work presents a new algorithm inspired by SC algorithm, the Co-Evolutionary Multi-Objective Genetic Graph based Clustering (CEMOG) algorithm, which is based on the Multi-Objective Genetic Graph-based Clustering (MOGGC) algorithm and extends it by introducing an adaptive number of clusters. CEMOG takes an island-model approach where each island keeps a population of candidate solutions for ki clusters. Individuals in the islands can migrate to encourage genetic diversity and the propagation of individuals around promising search regions. This new approach shows its competitive performance, compared to several classical clustering algorithms (EM, SC and K-means), through a set of experiments involving synthetic and real datasets. }} @InProceedings{Bidlo:2014:CEC, title = {Evolving Multiplication as Emergent Behavior in Cellular Automata Using Conditionally Matching Rules}, author = {Michal Bidlo}, pages = {2732--2739}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Developmental Systems}, abstract = { In this paper a special representation technique called conditionally matching rules will be applied in order to design computational processes in uniform cellular automata. The goal is to verify abilities of this approach in combination with genetic algorithm on the problem of designing various cellular automata that exhibit a given computational process. The principle of a computational process in a cellular automaton is to interpret some cells as input bits and some (possibly other) cells as output bits (i.e. the result of the computation). The genetic algorithm is applied to find a suitable transition function of a cellular automaton according to which the given computation could be observed during its development for all the possible binary combinations stored in the input cells. Both the input values and the result is represented by state values of cells. The input of the computation will be represented by the initial state of the cellular automaton. After a finite number of development steps the cells representing the output bits are expected to contain the result of the computation. A set of experiments will be performed considering various setups of the evolutionary system and arrangements of the target computation. It will be shown that non-trivial computations can be realised in a uniform two-dimensional cellular array. }} % Session: FrE2-4 Biometrics, Bioinformatics and Biomedical Applications @InProceedings{Menendez:2014:CECa, title = {Combining Graph Connectivity and Genetic Clustering to Improve Biomedical Summarization}, author = {Hector D. Menendez and Laura Plaza and David Camacho}, pages = {2740--2747}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Data mining, Biometrics, bioinformatics and biomedical applications, Classification, clustering and data analysis}, abstract = { Automatic summarisation is emerging as a feasible instrument to help biomedical researchers to access online literature and face information overload. The Natural Language Processing community is actively working toward the development of effective summarisation applications; however, they are sometimes less informative than the user needs. In this work, our aim is to improve a summarisation graph-based process combining genetic clustering with graph connectivity information. In this way, while genetic clustering allows us to identify the different topics that are dealt with in a document, connectivity information (in particular, degree centrality) allows us to assess and exploit the relevance of the different topics. Our automatic summaries are compared with others produced by commercial and research applications, to demonstrate the appropriateness of using this combination of techniques for automatic summarisation. }} @InProceedings{Datta:2014:CEC, title = {Selecting the Optimal {EEG} Electrode Positions for a Cognitive Task Using an Artificial Bee Colony with Adaptive Scale Factor Optimization Algorithm}, author = {Shreyasi Datta and Pratyusha Rakshit and Amit Konar and Atulya K. Nagar}, pages = {2748--2755}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Biometrics, bioinformatics and biomedical applications, Evolutionary programming, Numerical optimisation}, abstract = { The present work introduces a proposed Artificial Bee Colony with Adaptive Scale Factor (ABC-ASF) optimisation algorithm-based optimal electrode selection strategy from which the acquired EEG signals enlighten the major brain activities involved in a cognitive task. In ABC-ASF, the scale factor for mutation in traditional Artificial Bee Colony is self adapted by learning from the previous experiences. Experimental results obtained from the real framework of estimating optimal electrodes indicate that the proposed algorithm outperforms other state-of-art techniques with respect to computational accuracy and run-time complexity. }} @InProceedings{Ahmed:2014:CEC, title = {A New {GP}-Based Wrapper Feature Construction Approach to Classification and Biomarker Identification}, author = {Soha Ahmed and Mengjie Zhang and Lifeng Peng}, pages = {2756--2763}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Evolutionary programming, Biometrics, bioinformatics and biomedical applications}, abstract = { Mass spectrometry (MS) is a technology used for identification and quantification of proteins and metabolites. It helps in the discovery of proteomic or metabolomic biomarkers, which aid in diseases detection and drug discovery. The detection of biomarkers is performed through the classification of patients from healthy samples. The mass spectrometer produces high dimensional data where most of the features are irrelevant for classification. Therefore, feature reduction is needed before the classification of MS data can be done effectively. Feature construction can provide a means of dimensionality reduction and aims at improving the classification performance. In this paper, genetic programming (GP) is used for construction of multiple features. Two methods are proposed for this objective. The proposed methods work by wrapping a Random Forest (RF) classifier to GP to ensure the quality of the constructed features. Meanwhile, five other classifiers in addition to RF are used to test the impact of the constructed features on the performance of these classifiers. The results show that the proposed GP methods improved the performance of classification over using the original set of features in five MS data sets. }} @InProceedings{Byrne:2014:CEC, title = {An Examination of Synchronisation in Artificial Gene Regulatory Networks}, author = {Jonathan Byrne and Miguel Nicolau and Anthony Brabazon and Michael O'Neill}, pages = {2764--2769}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Developmental Systems, Complex Networks and Evolutionary Computation}, abstract = { An Artificial Genetic Regulatory Network (GRN) is a model of the gene expression regulation mechanism in biological organisms. It is a dynamical system that is capable of mimicking non-linear time series. The GRN was adapted to allow for input and output so that the system's rich dynamics could be used for dynamic problem solving. In order for the GRN to be embedded in the environment, the time scale of the physical system has to be mapped to that of the GRN and so a synchronisation process was introduced. This work examines the impact of different synchronisation intervals and how they effect the overall performance of the GRN. A variable synchronisation step that stops once the system has stabilised is also explored as a mechanism for automatically choosing the interval size. }} @InProceedings{Soncco-Alvarez:2014:CEC, title = {Memetic Algorithm for Sorting Unsigned Permutations by Reversals}, author = {Jose Luis Soncco-Alvarez and Mauricio Ayala-Rincon}, pages = {2770--2777}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic, multi-meme and hybrid algorithms, Genetic algorithms, Biometrics, bioinformatics and biomedical applications}, abstract = { Sorting by reversals unsigned permutations is a problem exhaustively studied in the fields of combinatorics of permutations and bioinformatics with crucial applications in the analysis of evolutionary distance between organisms. This problem was shown to be NP-hard, which gave rise to the development of a series of approximation and heuristic algorithms. Among these approaches, evolutionary algorithms were also proposed, from which to the best of our knowledge a parallel version of the first proposed genetic algorithm computes the highest quality results. These solutions were not optimised for the case when the population reaches a degenerate state, that is when individuals of the population remain very similar, and the procedure still continues consuming computational resources, but without improving the individuals. In this paper, a memetic algorithm is proposed for sorting unsigned permutations by reversals, using the local search as a way to improve the fitness function image of the individuals. Also, the entropy of the population is controlled, such that, when a degenerate state is reached the population is restarted. Several experiments were performed using permutations generated from biological data as well as hundreds of randomly generated permutations of different size, from which some ones were chosen and used as benchmark permutations. Experiments have shown that the proposed memetic algorithm uses more adequately the computational resources and gives competitive results in comparison with the parallel genetic algorithm and outperforms the results of the standard genetic algorithm. }} @InProceedings{Fogel:2014:CEC, title = {Evolved Neural Networks for {HIV-1} Co-Receptor Identification}, author = {Gary Fogel and Enoch Liu and Marco Salemi and Susanna Lamers and Michael McGrath}, pages = {2778--2784}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Computational Intelligence in Bioinformatics}, abstract = { HIV-1 infects a variety of cell types such as macrophages, T-cells and dendritic cells by expressing different chemokine receptors. R5 HIV-1 viruses use the CCR5 co-receptor for entry, X4 viruses use the CXCR4 co-receptor, and several viral strains make use of both co-receptors (a so-called "dual tropic" or R5X4 virus). Both X4 and R5X4 viruses are associated with late stage rapid progression to AIDS. It remains difficult to identify viral co-receptor type in advance of treatment, especially the R5X4 variety. In this paper we extended previous work to classify HIV-1 tropism using evolved neural networks and a larger set of HIV-1 sequences and features to improve overall classification accuracy. }} % Session: FrE2-5 Robotics and Engineering Applications @InProceedings{Di-Mario:2014:CEC, title = {Analysis of Fitness Noise in Particle Swarm Optimization: From Robotic Learning to Benchmark Functions}, author = {Ezequiel Di Mario and Inaki Navarro and Alcherio Martinoli}, pages = {2785--2792}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Robotics, Particle swarm optimisation (PSO), Dynamic and uncertain environments}, abstract = { Population-based learning techniques have been proved to be effective in dealing with noise and are thus promising tools for the optimisation of robotic controllers, which have inherently noisy performance evaluations. This article discusses how the results and guidelines derived from tests on benchmark functions can be extended to the fitness distributions encountered in robotic learning. We show that the large-amplitude noise found in robotic evaluations is disruptive to the initial phases of the learning process of PSO. Under these conditions, neither increasing the population size nor increasing the number of iterations are efficient strategies to improve the performance of the learning. We also show that PSO is more sensitive to good spurious evaluations of bad solutions than bad evaluations of good solutions, i.e., there is a non-symmetric effect of noise on the performance of the learning. }} @InProceedings{Pretorius:2014:CEC, title = {A Comparison of Neural Networks and Physics Models as Motion Simulators for Simple Robotic Evolution}, author = {Christiaan Pretorius and Mathys du Plessis and John Gonsalves}, pages = {2793--2800}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Robotics, Genetic algorithms, Evolutionary simulation-based optimisation}, abstract = { Robotic simulators are used extensively in Evolutionary Robotics (ER). Such simulators are typically constructed by considering the governing physics of the robotic system under investigation. Even though such physics-based simulators have seen wide usage in ER, there are some potential challenges involved in their construction and usage. An alternative approach to developing robotic simulators for use in ER, is to sample data directly from the robotic system and construct simulators based solely on this data. The authors have previously shown the viability of this approach by training Artificial Neural Networks (ANNs) to act as simulators in the ER process. It is, however, not known how this approach to simulator construction will compare to physics-based approaches, since a comparative study between ANN-based and physics-based robotic simulators in ER has not yet been conducted. This paper describes such a comparative study. Robotic simulators for the motion of a differentially-steered mobile robot were constructed using both ANN-based and physics-based approaches. These two approaches were then compared by employing each of the developed simulators in the ER process to evolve simple navigation controllers for the experimental robot in simulation. Results obtained indicated that, for the robotic system investigated in this study, ANN-based robotic simulators offer a promising alternative to physics-based simulators. }} @InProceedings{Moshaiov:2014:CEC, title = {Family Bootstrapping: A Genetic Transfer Learning Approach for Onsetting the Evolution for a Set of Related Robotic Tasks}, author = {Amiram Moshaiov and Amir Tal}, pages = {2801--2808}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Robotics, Convergence, scalability and complexity analysis}, abstract = { Studies on the bootstrap problem in evolutionary robotics help lifting the barrier from the way to evolve robots for complex tasks. It remains an open question, though, how to reduce the need for designer knowledge when devising a bootstrapping approach for any particular complex task. Transfer learning may help reducing this need and support the evolution of solutions to complex tasks, through task relatedness. Relying on the commonalities of similar tasks, we introduce a new concept of Family Bootstrapping (FB). FB refers to the creation of biased ancestors that are expected to onset the evolution of "a family" of solutions not just for one task, but for a set of related robot tasks. A general FB paradigm is outlined and the unique potential of the proposed concept is discussed. To highlight the validity of the FB concept, a simple demonstration case, concerning the evolution of neuro-controllers for a set of robot navigation tasks, is provided. The paper is concluded with some suggestions for future research. }} @InProceedings{Moshaiov:2014:CECa, title = {Is {MO-CMA-ES} Superior to {NSGA-II} for the Evolution of Multi-Objective Neuro-Controllers?}, author = {Amiram Moshaiov and Omer Abramovich}, pages = {2809--2816}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multi-objective evolutionary algorithms, Evolutionary Robotics, Evolved neural networks}, abstract = { In the last decade evolutionary multi-objective optimisers have been employed in studies concerning evolutionary robotics. In particular, the majority of such studies involve the evolution of neuro-controllers using either a genetic algorithm approach or an evolution strategies approach. Given the fundamental difference between these types of search mechanisms, a valid question is which kind of multi-objective optimiser is better for such applications. This question, which is dealt with here, is raised in view of the permutation problem that exists in evolutionary neural-networks. Two well-known Multi-objective Evolutionary Algorithms are used in the current comparison, namely MO-CMA-ES and NSGA-II. A multi-objective navigation problem is used for the testing, which is known to suffer from a local Pareto problem. For the employed simulation case MO-CMA-ES is better at finding a large sub-set of the approximated Pareto-optimal neuro-controllers, whereas NSGA-II is better at finding a complementary sub-set of the optimal controllers. This suggests that, if this phenomenon persists over a large range of case studies, then future studies should consider some modifications to such algorithms for the multi-objective evolution of neuro-controllers. }} @InProceedings{Dornberger:2014:CEC, title = {Optimization of the Picking Sequence of an Automated Storage and Retrieval System ({AS/RS})}, author = {Rolf Dornberger and Thomas Hanne and Remo Ryter and Stauffer Michael}, pages = {2817--2824}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Robotics, Engineering applications}, abstract = { In this paper we consider the problem of an optimal picking order sequence in a multi-aisle warehouse that is operated by a single automatic storage and retrieval system (AS/RS). The problem is solved by using a genetic algorithm (GA) similar to the one in the earlier research [3]. The problem and the solution approach are implemented in the OpenOpal software which provides a suitable test bed for simulation and optimisation (see http://www.openopal.org/). As a result it becomes evident that the genetic algorithm can be improved by changing the selection method and introducing an elitism mechanism. }} @InProceedings{Alam:2014:CEC, title = {Practical Application of an Evolutionary Algorithm for the Design and Construction of a Six-Inch Submarine}, author = {Khairul Alam and Tapabrata Ray and Sreenatha G. Anavatti}, pages = {2825--2832}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Engineering applications, Robotics}, abstract = { Unmanned underwater vehicles (UUVs) are becoming an attractive option for maritime search and survey operations as they are cheap and efficient compared to conventional use of divers or manned submersibles. Consequently, there has been a growing interest in UUV research among scientific and engineering communities. Although UUVs have received significant research interest in recent years, limited attention has been paid towards design and development of mini/micro UUVs (usually less than 1 foot in length). Micro UUVs are particularly attractive for deployment in extraordinarily confined spaces such as inspection of intricate underwater structures, ship wrecks, oil pipe lines or extreme hazardous areas. This paper considers previous work done in the field of miniature UUVs and presents an optimisation framework for preliminary design of that class of UUVs. A state-of-the-art optimisation algorithm namely infeasibility driven evolutionary algorithm (IDEA) is used to carry out optimisation of the micro UUV designs. The framework is subsequently used to identify optimal design of a torpedo-shaped micro UUV with an overall length of six inches (152.4 mm). The preliminary design identified through the process of optimisation is further analysed with the help of a computer-aided design tool to come up with a detailed design. The final design has since then been built and is currently undergoing trials. }} % Session: FrE3-1 Large-Scale Problems and Real-World Applications @InProceedings{Kazimipour:2014:CECb, title = {A Novel Hybridization of Opposition-Based Learning and Cooperative Co-Evolutionary for Large-Scale Optimization}, author = {Borhan Kazimipour and Mohammad Nabi Omidvar and Xiaodong Li and A.K. Qin}, pages = {2833--2840}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Recent Advances on Opposition-Based Learning \& Applications, Large-scale problems, Differential evolution}, abstract = { Opposition-based learning (OBL) and cooperative co-evolution (CC) have demonstrated promising performance when dealing with large-scale global optimisation (LSGO) problems. In this work, we propose a novel framework for hybridising these two techniques, and investigate the performance of simple implementations of this new framework using the most recent LSGO benchmarking test suite. The obtained results verify the effectiveness of our proposed OBL-CC framework. Moreover, some advanced statistical analyses reveal that the proposed hybridisation significantly outperforms its component methods in terms of the quality of finally obtained solutions. }} @InProceedings{Cooper:2014:CEC, title = {Optimising Large Scale Public Transport Network Design Problems Using Mixed-Mode Parallel Multi-Objective Evolutionary Algorithms}, author = {Ian Cooper and Matthew John and Rhydian Lewis and Andrew Olden and Christine Mumford}, pages = {2841--2848}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Parallel and distributed algorithms, Large-scale problems}, abstract = { In this paper we present a novel tool, using both OpenMP and MPI protocols, for optimising the efficiency of Urban Transportation Systems within a defined catchment, town or city. We build on a previously presented model which uses a Genetic Algorithm with novel genetic operators to optimise route sets and provide a transport network for a given problem set. This model is then implemented within a Parallel Multi-Objective Genetic Algorithm and demonstrated to be scalable to within the scope of real world, [city wide], problems. This paper compares and contrasts three methods of parallel distribution of the Genetic Algorithm's computational workload: a job farming algorithm and two variations on an `Islands' approach. Results are presented in the paper from both single and mixed mode strategies. The results presented are from a range of previously published academic problem sets. Additionally a real world inspired problem set is evaluated and a visualisation of the optimised output is given. }} @InProceedings{Watanabe:2014:CECa, title = {Many-Objective Evolutionary Computation for Optimization of Separated-Flow Control Using a {DBD} Plasma Actuator}, author = {Takeshi Watanabe and Tomoaki Tatsukawa and Antonio Lopez Jaimes and Hikaru Aono and Taku Nonomura and Akira Oyama and Kozo Fujii}, pages = {2849--2854}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Multi-objective evolutionary algorithms, Large-scale problems}, abstract = { In this paper, an algorithm for many-objective evolutionary computation, which is based on the NSGA-II with the Chebyshev preference relation, is applied to multi-objective design optimisation problem of dielectric barrier discharge plasma actuator (DBDPA). The present optimisation problem has four design parameters and six objective functions. The main goal of the paper is to extract useful design guidelines to predict control flow behaviour based on the DBDPA parameter values using the resulting approximation Pareto set obtained by the optimisation. }} @InProceedings{Lin:2014:CECa, title = {A Hybrid {EA} for High-Dimensional Subspace Clustering Problem}, author = {Lin Lin and Gen Mitsuo and Liang Yan}, pages = {2855--2860}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Genetic algorithms, Particle swarm optimisation (PSO), Discrete and combinatorial optimisation}, abstract = { Considering Particle Swarm Optimisation (PSO) could enhance solutions generated during the evolution process by exploiting their social knowledge and individual memory, we used PSO as a local search strategy in Genetic Algorithm (GA) framework for fine tuning the search space. GA is to make sure that every region of the search space is covered so that we have a reliable estimate of the global optimal solution and PSO is for further pruning the good solutions by searching around the neighbourhood. In this paper, proposed approach is used for subspace clustering, which is an extension of traditional clustering that seeks to find clustering in different subspaces within a dataset. Subspace clustering is to find a subset of dimensions on which to improve cluster quality by removing irrelevant and redundant dimensions in high dimensions problems. The experimental results demonstrate the positive effects of PSO as a local optimiser. }} @InProceedings{Du:2014:CECa, title = {A Simplified Glowworm Swarm Optimization Algorithm}, author = {Ming-yu Du and Xiu-juan Lei and Zhen-qiang Wu}, pages = {2861--2868}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation}, abstract = { Aimed at the poor optimising ability and the low accuracy of the glowworm swarm optimisation algorithm (GSO), a simplified glowworm swarm optimisation algorithm (SGSO) was put forward in this paper, which omitted the phases of seeking dynamic decision domain and movement probability calculation, and meanwhile simplified the location updating process. Moreover, elitism was introduced to improve the capacity of searching optimal solution. It was applied to the unimodal and multimodal benchmark function optimisation problems. The improved SGSO algorithm is compared with the basic GSO and other swarm intelligent optimisation algorithms to demonstrate the performance. Experimental results showed that SGSO improves not only the precision but also the efficiency in function optimisation. }} @InProceedings{Li:2014:CECp, title = {An Improved Two Archive Algorithm for Many-Objective Optimization}, author = {Bingdong Li and Jinlong Li and Ke Tang and Xin Yao}, pages = {2869--2876}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Large-scale problems, Multi-objective evolutionary algorithms, Multiobjective optimisation}, abstract = { Multi-Objective Evolutionary Algorithms have been deeply studied in the research community and widely used in the real-world applications. However, the performance of traditional Pareto-based MOEAs, such as NSGA-II and SPEA2, may deteriorate when tackling Many-Objective Problems, which refer to the problems with at least four objectives. The main cause for the degradation lies in that the high-proportional non-dominated solutions severely weaken the differentiation ability of Pareto-dominance. This may lead to stagnation. The Two Archive Algorithm (TAA) uses two archives, namely Convergence Archive (CA) and Diversity Archive (DA) as non-dominated solution repositories, focusing on convergence and diversity respectively. However, as the objective dimension increases, the size of CA increases enormously, leaving little space for DA. Besides, the update rate of CA is quite low, which causes severe problems for TAA to drive forth. Moreover, since TAA prefers DA members that are far away from CA, DA might drag the population backwards. In order to deal with these weaknesses, this paper proposes an improved version of TAA, namely ITAA. Compared to TAA, ITAA incorporates a ranking mechanism for updating CA which enables truncating CA while CA overflows. Besides, a shifted density estimation technique is embedded to replace the old ranking method in DA. The efficiency of ITAA is demonstrated by the experimental studies on benchmark problems with up to 20 objectives. }} % Session: FrE3-2 Evolvable Hardware and Software and Genetic Programming @InProceedings{Xiao:2014:CEC, title = {Two Step Evolution Strategy for Device Motif {BSIM} Model Parameter Extraction}, author = {Yang Xiao and Martin Trefzer and James Walker and Simon Bale and Andy Tyrrell}, pages = {2877--2884}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, EHW, Evolvable hardware and software}, abstract = { The modelling and simulation of semiconductor devices is a difficult and computationally intensive task. However the expense of fabrication and testing means that accurate modelling and simulation are crucial to the continued progress of the industry. To create these models and then perform the simulations requires parameters from accurate physical models to be obtained and then more abstract models created that can perform more complex circuit simulations. Device models (motifs) are created as a mitigation technique for improvement the circuit performance and as technology advances to help with the effects of transistor variability. In order to explore the characteristics of new device motifs on circuit designs, obtaining accurate and reliable device models becomes the first problem for designers. In this paper a Two Step Evolution Strategy (2SES) is proposed for device parameter model extraction. The proposed 2SES approach automatically extracts a set of parameters with respect to a specified device model. Compared with conventional mathematical extraction approach, 2SES is an efficient and accurate method to solve the parameter extraction problem and simultaneously addresses the fact of the mathematical extraction having the complexity of Multi-objective optimisation. Compared with single step ES extract result, it is shown that the two-step ES extraction process continues improving generations by adjusting the optimisation parameters. Finally, an application of a new device motif on circuit design is given at end of the paper and compared against a standard device. }} @InProceedings{Wagner:2014:CECa, title = {Maximising Axiomatization Coverage and Minimizing Regression Testing Time}, author = {Markus Wagner}, pages = {2885--2892}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Evolvable hardware and software}, abstract = { The correctness of program verification systems is of great importance, as they are used to formally prove that safety and security critical programs follow their specification. One of the contributing factors to the correctness of the whole verification system is the correctness of the background axiomatisation, which captures the semantics of the target program language. We present a framework for the maximisation of the proportion of the axiomatisation that is used ("covered") during testing of the verification tool. The diverse set of test cases found not only increases the trust in the verification system, but it can also be used to reduce the time needed for regression testing. }} @InProceedings{Huo:2014:CEC, title = {A New Adaptive Kalman Filter by Combining Evolutionary Algorithm and Fuzzy Inference System}, author = {Yudan Huo and Zhihua Cai and Wenyin Gong and Qin Liu}, pages = {2893--2900}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {EC Generalisation in High-dimensional Input Spaces, Engineering applications, Single Objective Numerical Optimisation}, abstract = { The performance of the Kalman filter (KF), which is recognised as an outstanding tool for dynamic system state estimation, heavily depends on its parameter R, called the measurement noise covariance matrix. However, it's difficult to get the exact value of R before the filter starts, and the value of R is likely to change with the measurement environment when the filter is working. To solve this problem, a new parameter adaptive Kalman filter is proposed in this paper. In this new Kalman filter, the initial value of R is offline decided by Evolutionary Algorithm (EA), and the value of R decided by EA is online updated by Fuzzy Inference System (FIS). A simulation experiment based on target tracking is carried out, and the results demonstrate that the new adaptive Kalman filter proposed in this paper (HydGeFuzKF) has a stronger adaptability to time-varying measurement noises than regular Kalman filter (RegularKF), Sage-Husa adaptive Kalman filter (SageHusaKF), the adaptive Kalman filter only based on genetic algorithm (GeneticKF) and the adaptive Kalman filter only based on fuzzy inference system (FuzzyKF). }} @InProceedings{Sekanina:2014:CEC, title = {Cartesian Genetic Programming as Local Optimizer of Logic Networks}, author = {Lukas Sekanina and Ondrej Ptak and Zdenek Vasicek}, pages = {2901--2908}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Cartesian genetic programming, Hardware Aspects of Bio-Inspired Architectures and Systems (HABIAS)}, abstract = { Logic synthesis and optimisation methods work either globally on the whole logic network or locally on preselected sub-networks. Evolutionary design methods have already been applied to evolve and optimise logic circuits at the global level. In this paper, we propose a new method based on Cartesian genetic programming (CGP) as a local area optimiser in combinational logic networks. First, a subcircuit is extracted from a complex circuit, then the subcircuit is optimised by CGP and finally the optimised subcircuit replaces the original one. The procedure is repeated until a termination criterion is satisfied. We present a performance comparison of local and global evolutionary optimisation methods with a conventional approach based on ABC and analyse these methods using differently pre-optimised benchmark circuits. If a sufficient time is available, the proposed locally optimising CGP gives better results than other locally operating methods reported in the literature; however, its performance is significantly worse than the evolutionary global optimisation. }} @InProceedings{Donne:2014:CEC, title = {Wave Height Quantification Using Land Based Seismic Data with Grammatical Evolution}, author = {Sarah Donne and Miguel Nicolau and Christopher Bean and Michael O'Neill}, pages = {2909--2916}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Grammatical Evolution, Real-world applications}, abstract = { Accurate, real time, continuous ocean wave height measurements are required for the initialisation of ocean wave forecast models, model hindcasting, and climate studies. These measurements are usually obtained using in situ ocean buoys or by satellite altimetry, but are sometimes incomplete due to instrument failure or routine network upgrades. In such situations, a reliable gap filling technique is desirable to provide a continuous and accurate ocean wave field record. Recorded on a land based seismic network are continuous seismic signals known as microseisms. These microseisms are generated by the interactions of ocean waves and will be used in the estimation of ocean wave heights. Grammatical Evolution is applied in this study to generate symbolic models that best estimate ocean wave height from terrestrial seismic data, and the best model is validated against an Artificial Neural Network. Both models are tested over a five month period of 2013, and an analysis of the results obtained indicates that the approach is robust and that it is possible to estimate ocean wave heights from land based seismic data. }} @InProceedings{Xie:2014:CECa, title = {Genetic Programming Based Activity Recognition on a Smartphone Sensory Data Benchmark}, author = {Feng Xie and Andy Song and Vic Ciesielski}, pages = {2917--2924}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, Genetic programming, Classification, clustering and data analysis, Real-world applications}, abstract = { Activity recognition from smart phone sensor inputs is of great importance to enhance user experience. Our study aims to investigate the applicability of Genetic Programming (GP) approach on this complex real world problem. Traditional methods often require substantial human efforts to define good features. Moreover the optimal features for one type of activity may not be suitable for another. In comparison, our GP approach does not require such feature extraction process, hence, more suitable for complex activities where good features are difficult to be pre-defined. To facilitate this study we therefore propose a benchmark of activity data collected from various smartphone sensors, as currently there is no existing publicly available database for activity recognition. In this study, a GP-based approach is applied to nine types of activity recognition tasks by directly taking raw data instead of features. The effectiveness of this approach can be seen by the promising results. In addition our benchmark data provides a platform for other machine learning algorithms to evaluate their performance on activity recognition. }} % Session: FrE3-3 Swarm Intelligence @InProceedings{Janecek:2014:CEC, title = {Swarm/Evolutionary Intelligence for Agent-Based Social Simulation}, author = {Andreas Janecek and Tobias Jordan and Fernando Buarque de Lima-Neto}, pages = {2925--2932}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Finance and economics, Evolutionary simulation-based optimisation, Particle swarm optimisation (PSO)}, abstract = { Several micro economic models allow to evaluate consumer's behaviour using a utility function that is able to measure the success of an individual's decision. Such a decision may consist of a tuple of goods an individual would like to buy and hours of work necessary to pay for them. The utility of such a decision depends not only on purchase and consumption of goods, but also on fringe benefits such as leisure, which additionally increases the utility to the individual. Utility can be used then as a collective measure for the overall evaluation of societies. In this paper, we present and compare three different agent based social simulations in which the decision finding process of consumers is performed by three algorithms from swarm intelligence and evolutionary computation. Although all algorithms appear to be suitable for the underlying problem as they are based on historical information and also contain a stochastic part which allows for modelling the uncertainty and bounded rationality, they differ greatly in terms of incorporating historical information used for finding new alternative decisions. Newly created decisions that violate underlying budget constraints may either be mapped back to the feasible region, or may be allowed to leave the valid search space. However, in order to avoid biases that would disrupt the inner rationale of each meta heuristic, such invalid decisions are not remembered in the future. Experiments indicate that the choice of such bounding strategy varies according to the choice of the optimisation algorithm. Moreover, it seems that each of the techniques could excel in identifying different types of individual behaviour such as risk affine, cautious and balanced. }} @InProceedings{Zan:2014:CEC, title = {Solving the Multidimensional Knapsack Problem Using a {CUDA} Accelerated {PSO}}, author = {Drahoslav Zan and Jiri Jaros}, pages = {2933--2939}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Discrete and combinatorial optimisation}, abstract = { The Multidimensional Knapsack Problem (MKP) represents an important model having numerous applications in combinatorial optimisation, decision-making and scheduling processes, cryptography, etc. Although the MKP is easy to define and implement, the time complexity of finding a good solution grows exponentially with the problem size. Therefore, novel software techniques and hardware platforms are being developed and employed to reduce the computation time. This paper addresses the possibility of solving the MKP using a GPU accelerated Particle Swarm Optimisation (PSO). The goal is to evaluate the attainable performance benefit when using a highly optimised GPU code instead of an efficient multi-core CPU implementation, while preserving the quality of the search process. The paper shows that a single Nvidia GTX 580 graphics card can outperform a quad-core CPU by a factor of 3.5 to 9.6, depending on the problem size. As both implementations are memory bound, these speed-ups directly correspond to the memory bandwidth ratio between the investigated GPU and CPU. }} @InProceedings{Runkler:2014:CEC, title = {Multidimensional Scaling with Multiswarming}, author = {Thomas Runkler and James Bezdek}, pages = {2940--2946}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Classification, clustering and data analysis, Particle swarm optimisation (PSO), Data mining}, abstract = { We introduce a new method for multidimensional scaling in dissimilarity data that is based on preservation of metric topology between the original and derived data sets. The model seeks neighbours in the derived data that have the same ranks as in the input data. The algorithm we use to optimise the model is a modification of particle swarm optimisation called multiswarming. We compare the new method to three well known approaches: Principal component analysis, Sammon's method, and (Kruskal's) metric MDS. Our method produces feature vector realisations that compare favourably with the other approaches on three real relational data sets. }} @InProceedings{Metlicka:2014:CEC, title = {Chaos-Driven Discrete Artificial Bee Colony}, author = {Magdalena Metlicka and Donald Davendra}, pages = {2947--2954}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computing with Deterministic Chaos}, abstract = { In this paper, a chaos driven Discrete Artificial Bee Algorithm is introduced. The main premise of this work is to ascertain if using chaos maps in lieu of standard pseudorandom number generators can improve the performance of the canonical algorithm. Nine unique chaos maps are embedded in the Discrete Artificial Bee Algorithm alongside the Mersenne twister and evaluated on the lot-streaming flowshop scheduling problem with setup time. Based on the obtained results, a number of chaotic maps significantly improve the performance of the algorithm. Additionally, the new algorithm is favourably compared with the chaos driven Enhanced Differential Evolution algorithm for the same problem. }} @InProceedings{Alam:2014:CECa, title = {Web Bots Detection Using Particle Swarm Optimization Based Clustering}, author = {Shafiq Alam and Gillian Dobbie and Yun Sing Koh and Patricia Riddle}, pages = {2955--2962}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Data mining, Real-world applications}, abstract = { Optimisation based techniques have emerged as important methods to tackle the problems of efficiency and accuracy in data mining. One of the current application areas is outlier detection which has not been fully explored yet but has enormous potential. Web bots are an example of outliers, which can be found in the web usage analysis process. Web bot requests are different from a genuine web user as web bots crawl large numbers of pages in a very short time. If web bots remains undetected they can skew the analysis process which can result in incorrect patterns that can cause wrong decisions. In this paper we use one of the popular Swarm Intelligence (SI) based techniques called Particle Swarm Optimisation (PSO) to detect web bots among genuine users request. We use our Particle Swarm Optimisation (PSO) based clustering algorithm, Hierarchical Particle Swarm Optimisation based clustering (HPSO-clustering) to cluster the web usage data and detect the abnormal behaviour caused by the web bots. We present the results of detection which shows that our proposed approach is capable of detecting such abnormal behaviour. We then compare the characteristics of the detected web bots with genuine web-users using cross validation. }} @InProceedings{Wu:2014:CECf, title = {An Ant Colony Optimization Algorithm for Multi-Objective Clustering in Mobile Ad Hoc Networks}, author = {Chung-Wei Wu and Tsung-Che Chiang and Li-Chen Fu}, pages = {2963--2968}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent Network Systems}, abstract = { Due to the proliferation of smart mobile devices and the developments in wireless communication, mobile ad hoc networks (MANETs) are gaining more and more attention in recent years. Routing in MANETs is a challenge, especially when the network contains a large number of nodes. The clustering technique is a popular method to organise the nodes in MANETs. It divides the network into several clusters and assigns a cluster head to each cluster for intra and inter-cluster communication. Clustering is NP-hard and needs to consider multiple objectives. In this paper we propose a Pareto-based ant colony optimisation (ACO) algorithm to deal with this multiobjective optimisation problem. A new encoding scheme is proposed to reduce the size of search space, and a new decoding scheme is proposed to generate high-quality solutions effectively. Experimental results show that our approach is better than several benchmark approaches. }} % Session: FrE3-4 Heuristics, Metaheuristics and Hyper-Heuristics II @InProceedings{Adriaensen:2014:CEC, title = {Designing Reusable Metaheuristic Methods: A Semi-Automated Approach}, author = {Steven Adriaensen and Tim Brys and Ann Nowe}, pages = {2969--2976}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics}, abstract = { Many interesting optimisation problems cannot be solved efficiently. Recently, a lot of work has been done on meta heuristic optimisation methods that quickly find approximate solutions to otherwise intractable problems. While successful, the field suffers from a notable lack of reuse of methods, both in practical applications as in research. In this paper, we describe a semi-automated approach to design more re-usable methods, based on key principles of re-usability such as simplicity, modularity and generality. We illustrate this methodology by designing general metaheuristics (using hyperheuristics) and show that the methods obtained are competitive with the contestants of the Cross-Domain Heuristic Search Competition (2011). In particular, we find a method performing better than the competition's winner, which can be considered the state-of-the-art in domain-independent metaheuristic search. }} @InProceedings{Enaya:2014:CEC, title = {Network Path Optimization Under Dynamic Conditions}, author = {Yaser Enaya and Kalyanmoy Deb}, pages = {2977--2984}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristics, metaheuristics and hyper-heuristics, Numerical optimisation, Dynamic and uncertain environments}, abstract = { Most network optimisation problems are studied under a static scenario in which connectivity of the network and weights associated with the links of the networks are assumed to be fixed. However, in practice, they are likely to change with time and if the network is to be used over time under dynamic conditions, they need to be re-optimised as soon as there is a change. Since optimisation process requires some finite time, there is a need for a efficient dynamic optimisation strategy for solving such problems. In this study, we extend a previously proposed "Frozen-time" algorithm to network optimisation by which new and optimised networks can be obtained in a computationally fast manner. We propose three different variations of the optimisation strategies and show proof-of-principle simulation results on a 20-node network having 190 different source-destination paths. The results are interesting and suggest a viable further research. }} @InProceedings{Brent:2014:CEC, title = {A Parallel {Lagrangian-ACO} Heuristic for Project Scheduling}, author = {Oswyn Brent and Dhananjay Thiruvady and Antonio Gomez-Iglesias and Rodolfo Garcia-Flores}, pages = {2985--2991}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Parallel and distributed algorithms, Heuristics, metaheuristics and hyper-heuristics}, abstract = { In this paper we present a parallel implementation of an existing Lagrangian heuristic for solving a project scheduling problem. The original implementation uses Lagrangian relaxation to generate useful upper bounds and guidance towards generating good lower bounds or feasible solutions. These solutions are further improved with Ant Colony Optimisation via loose and tight couplings. While this approach has proved to be effective, there are often large gaps for a number of the problem instances. Thus, we aim to improve the performance of this algorithm through a parallel implementation on a multicore shared memory architecture. However, the original algorithm is inherently sequential and is not trivially parallelisable due to the dependencies between the different components involved. Hence, we propose different approaches to carry out this parallelisation. Our results show that the parallel version produces consistently better results given the same time limits. }} @InProceedings{Masi:2014:CEC, title = {A Multidirectional {Physarum} Solver for the Automated Design of Space Trajectories}, author = {Luca Masi and Massimiliano Vasile}, pages = {2992--2999}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Discrete and combinatorial optimisation, Ant colony optimisation}, abstract = { This paper proposes a bio-inspired algorithm to automatically generate optimal multi-gravity assist trajectories. The multi-gravity assist problem has some analogies with the better known Travellings Salesman Problem and can be addressed with similar strategies. An algorithm drawing inspiration from the Physarum slime mould is proposed to grow and explore a tree of decisions that corresponds to the possible sequences of transfers from one planet to another. Some examples show that the proposed bio-inspired algorithm can produce solutions that are better than the ones generated by humans or with Hidden Genes Genetic Algorithms. }} @InProceedings{Xie:2014:CECb, title = {A Genetic Programming-Based Hyper-heuristic Approach for Storage Location Assignment Problem}, author = {Jing Xie and Yi Mei and Andreas Ernst and Xiaodong Li and Andy Song}, pages = {3000--3007}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {genetic algorithms, genetic programming, Real-world applications}, abstract = { This study proposes a method for solving real-world warehouse Storage Location Assignment Problem (SLAP) under grouping constraints by Genetic Programming (GP). Integer Linear Programming (ILP) formulation is used to define the problem. By the proposed GP method, a subset of the items is repeatedly selected and placed into the available current best location of the shelves in the warehouse, until all the items have been assigned with locations. A heuristic matching function is evolved by GP to guide the selection of the subsets of items. Our comparison between the proposed GP approach and the traditional ILP approach shows that GP can obtain near-optimal solutions on the training data within a short period of time. Moreover, the evolved heuristics can achieve good optimisation results on unseen scenarios, comparable to that on the scenario used for training. This shows that the evolved heuristics have good reusability and can be directly applied for slightly different scenarios without any new search process. }} @InProceedings{Burman:2014:CEC, title = {The Monarchy Driven Optimization Algorithm}, author = {Ritambhar Burman and Swagatam Das and Zheshanul Haque and Athanasios V. Vasilakos and Soumyadip Chakraborti}, pages = {3008--3015}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Heuristics, metaheuristics and hyper-heuristics}, abstract = { We present a novel human society inspired algorithm for single-objective bound constrained optimisation. The proposed Monarchy Driven Optimisation (MDO) algorithm is a population-based iterative global optimisation technique for multi-dimensional and multi-modal problems. At its core, this technique introduces a monarchial society where the outlook of its population is fashioned by the thoughts of individuals and the monarch. A detailed study including the tuning of MDO parameters is presented along with the theory. It is applied to standard benchmark functions comprising uni-modal and multi-modal as well as rotated functions. The results section suggests that, in most instances, MDO outperforms other well-known techniques such as Particle Swarm Optimisation (PSO), Differential Evolution (DE), Gravitational Search Algorithm (GSA) and Comprehensive Learning Particle Swarm Optimisation(CLPSO) in terms of final convergence value and mean convergence value, thus proves to be a robust optimisation technique. }} % Session: FrE3-5 Real-World Applications II @InProceedings{Jin:2014:CEC, title = {Heuristic Optimization for Software Project Management with Impacts of Team Efficiency}, author = {Nanlin Jin and Xin Yao}, pages = {3016--3023}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {SBSE, Real-world applications, Evolutionary Multi-Objective Optimisation and Decision-Making}, abstract = { Most of the studies on project scheduling problems assume that every assigned participant or every team of the same number of participants, completes tasks with an equal efficiency, but this is usually not the case for real world problems. This paper presents a more realistic and complex model with extra considerations on team efficiency which are quantitatively measured on employee task assignment. This study demonstrates the impacts of team efficiency in a well-studied software project management problem. Moreover, this study illustrates how a heuristic optimisation method, population-based incremental learning, copes with such added complexity. The experimental results show that the resulting near optimal solutions not only satisfy constraints, but also reflect the impacts of team efficiency. The findings will hopefully motivate future studies on comprehensive understandings of the quality and efficiency of team work. }} @InProceedings{Wang:2014:CECk, title = {A Multiobjective Optimization Method Based on {MOEA/D} and Fuzzy Clustering for Change Detection in {SAR} Images}, author = {Qiao Wang and Hao Li and Maoguo Gong and Linzhi Su and Licheng Jiao}, pages = {3024--3029}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Multi-objective evolutionary algorithms}, abstract = { For the presence of speckle noise in SAR images, many change detection methods have been developed to suppress the effect of noise. However, all these methods will result in the loss of image details, and the trade-off between detail preserving and noise removing capability has become an urgent problem remaining to be settled. In this paper, we put forward an innovation for change detection in synthetic aperture radar images. It integrates evolutionary computation into fuzzy clustering process, and considers detail preserving capability and noise removing capability as two separate objectives for multiobjective optimisation, and thus transforming the change detection problem into a multiobjective optimisation problem (MOP). Experiments conducted on real SAR images confirm that the new approach is efficient. }} @InProceedings{Tsai:2014:CECa, title = {A Novel Evaluation Function for {LT} Codes Degree Distribution Optimization}, author = {Pei-Chuan Tsai and Chih-Ming Chen and Ying-ping Chen}, pages = {3030--3035}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Real-world applications, Engineering applications, Surrogate-assisted Global Optimisation Methods for Expensive Engineering Design}, abstract = { Luby transform (LT) codes implements an important property called ratelessness, meaning a fixed code rate is unnecessary and LT codes can complete the transmission without channel status. The property is advantageous to transmit over certain environments such as broadcasting in heterogeneous networks or transmitting data over unknown channels. For this reason, improving LT codes is a crucial research issue in recent years. The performance of LT codes is decided by the code length and a probability mass function, called degree distribution, used in the encoding process. To improve the performance of LT codes, many studies proposed to optimise the degree distribution by using methods in evolutionary computation. One of the key steps in the evolutionary process is to evaluate decision variables for comparing the fitness of each individual. In the optimisation of LT codes, it needs to repeatedly simulate the encoding/decoding process with a given distribution and evaluate the performance over a sufficient number of runs. Hence, a lot of computational resource is necessary for the optimisation of LT codes. In this paper, we propose a heuristic function to evaluate the performance of LT codes. The evaluation function estimates the expected fraction of unsolved symbols with the specified code length, reception overhead, and degree distribution. Based on the proposed function, a huge number of evaluations is possible for searching for better degree distributions. We first verify the practicality of the proposed function and then employ it in a multi-objective evolutionary algorithm to investigate the trade-off of LT codes between the computational cost and decoding performance. }} @InProceedings{Triguero:2014:CEC, title = {A Combined {MapReduce}-Windowing Two-Level Parallel Scheme for Evolutionary Prototype Generation}, author = {Isaac Triguero and Daniel Peralta and Jaume Bacardit and Salvador Garcia and Francisco Herrera}, pages = {3036--3043}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Computation in Big Data}, abstract = { Evolutionary prototype generation techniques have demonstrated their usefulness to improve the capabilities of the nearest neighbour classifier. They act as data reduction algorithms by generating representative points of a given problem. Their main purposes are to speed up the classification process and to reduce the storage requirements and sensitivity to noise of the nearest neighbour rule. Nowadays, with the increment of available data, the use of this kind of reduction techniques becomes more important. However, their applicability can be limited to problems with no more than tens of thousands of instances. In order to address this limitation, in this work we develop a two-level parallelisation scheme for evolutionary prototype generation methods. Firstly, it distributes the functioning of these algorithms in several tasks based on a MapReduce framework. Then, for each one of these tasks (mappers), we accelerate the prototype generation process by using a windowing approach. This model enables evolutionary prototype generation algorithms to be applied over large-scale classification problems without accuracy loss. Our preliminary experiments using a dataset of 1 million instances show that this proposal is an appropriate tool to improve the performance of the nearest neighbour classifier with big data. }} @InProceedings{Gu:2014:CECa, title = {A Dynamic-Weighted Collaborative Filtering Approach to Address Sparsity and Adaptivity Issues}, author = {Liang Gu and Peng Yang and Yongqiang Dong}, pages = {3044--3050}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent systems applications, Real-world applications, Data mining}, abstract = { Recommendation systems, as efficient measures to handle the information overload and personalised service problems, have attracted considerable attention in research community. Collaborative filtering is one of the most successful techniques based on the user-item matrix in recommendation systems. Usually the matrix is extremely sparse due to the massive number of users and items. And the sparsity of users and items tends to differ significantly in degree. The feature of the matrix changes with the variation of users/items data and hence, leads to poor scalability of the recommendation method. This paper proposes a dynamic weighted collaborative filtering approach (DWCF) to address sparsity and adaptivity issues. In this approach, the relationship between the distributions of similar users and items is considered to get better recommendation, i.e., the contributions of the user part and the item part to recommendation results depend on their similarity ratios. Moreover, the effect strength of different parts is controlled by an averaging parameter. Experiments on MovieLens dataset illustrate that the DWCF approach proposed in this paper can obtain good recommendation result given different conditions of data sparsity and perform better than a user-based predictor, an item-based predictor and a conventional hybrid approach. }} @InProceedings{Reid:2014:CEC, title = {Carry Trade Portfolio Optimization using Particle Swarm Optimization}, author = {Stuart Reid and Katherine Malan and Andries Engelbrecht}, pages = {3051--3058}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Finance and economics}, abstract = { Portfolio optimisation has as its objective to find optimal portfolios, which apportion capital between their constituent assets such that the portfolio's risk adjusted return is maximised. Portfolio optimisation becomes more complex as constraints are imposed, multiple sources of return are included, and alternative measures of risk are used. Meta-heuristic portfolio optimisation can be used as an alternative to deterministic approaches under increased complexity conditions. This paper uses a particle swarm optimisation (PSO) algorithm to optimise a diversified portfolio of carry trades. In a carry trade, investors profit by borrowing low interest rate currencies and lending high interest rate currencies, thereby generating return through the interest rate differential. However, carry trades are risky because of their exposure to foreign exchange losses. Previous studies showed that diversification does significantly mitigate this risk. This paper goes one step further and shows that meta-heuristic portfolio optimisation can further improve the risk adjusted returns of diversified carry trade portfolios. }} % Session: FrE4-1 Constraint-Handling and Preference-Handling @InProceedings{Bonyadi:2014:CEC, title = {On the Edge of Feasibility: A Case Study of the Particle Swarm Optimizer}, author = {Mohammad reza Bonyadi and Zbigniew Michalewicz}, pages = {3059--3066}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling, Particle swarm optimisation (PSO)}, abstract = { In many real-world constrained optimisation problems (COPs) it is highly probable that some constraints are active at optimum points, i.e. some optimum points are boundary points between feasible and infeasible parts of the search space. A method is proposed which narrows the feasible area of a COP to its boundary. In the proposed method the thickness of the narrowed boundary is adjustable by a parameter. The method is extended in a way that it is able to limit the feasible regions to boundaries where at least one of the constraints in a given subset of all constraints is active and the remaining constraints might be active or not. Another extension is able to limit the search to cases where all constraints in a given subset are active and the rest might be active or not. The particle swarm optimisation algorithm is used as a framework to compare the proposed methods. Results show that the proposed methods can limit the search to the requested boundary and they are effective in locating optimal solutions on the boundaries of the feasible and infeasible area. }} @InProceedings{Dong:2014:CECa, title = {Linear Sparse Arrays Designed by Dynamic Constrained Multi-Objective Evolutionary Algorithm}, author = {Wei Dong and Sanyou Zeng}, pages = {3067--3072}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Constraint handling}, abstract = { The design of linear sparse array is a constrained multi-objective optimisation problem(CMOP). There are three objectives: minimisation of peak side lobe level (PSLL), half-power beam width(HPBW) and spatial aperture. And the amplitude coefficients of elements and the sensor positions of the array are decision variables. Dynamic constrained multi-objective evolutionary algorithm(DCMOEA) is used to design linear sparse arrays in this paper. It makes a difference that the output is a set of Pareto solutions (antenna arrays), not just only one solution. The users can choose an array from the set to meet their preferences for low PSLL, small HPBW, small spatial aperture or a trade-off among them. Experimental results showed that the DCMOEA performs better than peer state-of-art algorithms referred in this paper, especially on the arrays' spatial aperture optimisation. }} @InProceedings{Si:2014:CEC, title = {Mapping Constrained Optimization Problems to Penalty Parameters: An Empirical Study}, author = {Chengyong Si and Jianqiang Shen and Xuan Zou and Lei Wang and Qidi Wu}, pages = {3073--3079}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Constraint handling, Evolutionary simulation-based optimisation}, abstract = { Penalty function method is one of the most popular used Constraint Handling Technique for Evolutionary Algorithms (EAs) solution selecting, whose performance is mainly determined by penalty parameters. This paper tries to study the penalty parameter from the aspect of problem characteristics, i.e., to construct a corresponding relationship between the problems and the penalty parameters. The experimental results confirm the relationship, which provides valuable reference for future algorithm design. }} @InProceedings{Singh:2014:CECb, title = {A Constrained Multi-Objective Surrogate-Based Optimization Algorithm}, author = {Prashant Singh and Ivo Couckuyt and Francesco Ferranti and Tom Dhaene}, pages = {3080--3087}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Surrogate-assisted Global Optimisation Methods for Expensive Engineering Design}, abstract = { Surrogate models or metamodels are widely used in the realm of engineering for design optimisation to minimise the number of computationally expensive simulations. Most practical problems often have conflicting objectives, which lead to a number of competing solutions which form a Pareto front. Multi-objective surrogate-based constrained optimisation algorithms have been proposed in literature, but handling constraints directly is a relatively new research area. Most algorithms proposed to directly deal with multi-objective optimisation have been evolutionary algorithms (Multi-Objective Evolutionary Algorithms MOEAs). MOEAs can handle large design spaces but require a large number of simulations, which might be infeasible in practice, especially if the constraints are expensive. A multi-objective constrained optimisation algorithm is presented in this paper which makes use of Kriging models, in conjunction with multi-objective probability of improvement (PoI) and probability of feasibility (PoF) criteria to drive the sample selection process economically. The efficacy of the proposed algorithm is demonstrated on an analytical benchmark function, and the algorithm is then used to solve a microwave filter design optimisation problem. }} @InProceedings{Poursoltan:2014:CEC, title = {A Feature-Based Analysis on the Impact of Linear Constraints for e-Constrained Differential Evolution}, author = {Shayan Poursoltan and Frank Neumann}, pages = {3088--3095}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Numerical optimisation, Differential evolution, Evolutionary computation theory}, abstract = { Feature-based analysis has provided new insights into what characteristics make a problem hard or easy for a given algorithms. Studies, so far, considered unconstrained continuous optimisation problem and classical combinatorial optimisation problems such as the Travelling Salesperson problem. In this paper, we present a first feature-based analysis for constrained continuous optimisation. To start the feature-based analysis of constrained continuous optimisation, we examine how linear constraints can influence the optimisation behaviour of the well-known e-constrained differential evolution algorithm. Evolving the coefficients of a linear constraint, we show that even the type of one linear constraint can make a difference of 10-30\% in terms of function evaluations for well-known continuous benchmark functions. }} @InProceedings{Ki-Baek:2014:CEC, title = {{DMOPSO}: Dual Multi-Objective Particle Swarm Optimization}, author = {Lee Ki-Baek and Kim Jong-Hwan}, pages = {3096--3102}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Multiobjective optimisation, Particle swarm optimisation (PSO), Preference handling}, abstract = { Since multi-objective optimisation algorithms (MOEAs) have to find exponentially increasing number of nondominated solutions with the increasing number of objectives, it is necessary to discriminate more meaningful ones from the other nondominated solutions by additionally incorporating user preference into the algorithms. This paper proposes dual multi-objective particle swarm optimisation (DMOSPO) by introducing secondary objectives of maximising both user preference and diversity to the nondominated solutions obtained for primary objectives. The proposed DMOSPO can induce the balanced exploration of the particles in terms of user preference and diversity through the dual-stage of nondominated sorting such that it can generate preferable and diverse nondominated solutions. To demonstrate the effectiveness of the proposed DMOPSO, empirical comparisons with other state-of-the-art algorithms are carried out for benchmark functions. Experimental results show that DMOPSO is competitive with the other compared algorithms and properly reflects the user's preference in the optimisation process while maintaining the diversity and solution quality. }} % Session: FrE4-2 Particle Swarm Optimisation @InProceedings{Cheng:2014:CECa, title = {Demonstrator Selection in a Social Learning Particle Swarm Optimizer}, author = {Ran Cheng and Yaochu Jin}, pages = {3103--3110}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Numerical optimisation}, abstract = { Social learning plays an important role in behaviour learning among social animals. Different from individual (asocial) learning, social learning has the advantage of allowing individuals to learn behaviours from others without the extra costs of individual trial-and-error. Inspired by the natural social learning phenomenon, we have transplanted the social learning mechanism into particle swarm optimisation (PSO) to develop a social learning PSO (SL-PSO). Unlike classical PSO variants, the SL-PSO is performed on a sorted swarm, and instead of merely learning from historical best positions, the particles are able to learn from anyone better (demonstrators) in the current swarm. A key mechanism in the SL-PSO is the learning strategy, where an imitator will learn from different demonstrators. However, in our previous work, little discussion has been focused on demonstrator selection, i.e., which demonstrators are to learn from by the imitator. In this paper, based on the analysis of the demonstrator selection in the SL-PSO, two demonstrator selection strategies are proposed. Experimental results show that, the proposed demonstrator selection strategies have significantly enhanced the performance of the SL-PSO in comparison to five representative PSO variants on a set of benchmark problems. }} @InProceedings{Nguyen:2014:CECc, title = {Filter Based Backward Elimination in Wrapper Based {PSO} for Feature Selection in Classification}, author = {Bach Hoai Nguyen and Bing Xue and Ivy Liu and Mengjie Zhang}, pages = {3111--3118}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Classification, clustering and data analysis, Data mining}, abstract = { The advances in data collection increase the dimensionality of the data (i.e. the total number of features) in many fields, which arises a challenge to many existing feature selection approaches. This paper develops a new feature selection approach based on particle swarm optimisation (PSO) and a local search that mimics the typical backward elimination feature selection method. The proposed algorithm uses a wrapper based fitness function, i.e. the classification error rate. The local search is performed only on the global best and uses a filter based measure, which aims to take the advantages of both filter and wrapper approaches. The proposed approach is tested and compared with three recent PSO based feature selection algorithms and two typical traditional feature selection methods. Experiments on eight benchmark datasets show that the proposed algorithm can be successfully used to select a significantly smaller number of features and simultaneously improve the classification performance over using all features. The proposed approach outperforms the three PSO based algorithms and the two traditional methods. }} @InProceedings{Xue:2014:CEC, title = {An Archive Based Particle Swarm Optimisation for Feature Selection in Classification}, author = {Bing Xue and A. K. Qin and Mengjie Zhang}, pages = {3119--3126}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Classification, clustering and data analysis, Data mining}, abstract = { Feature selection aims to select a subset of relevant features from typically a large number of original features, which is a difficult task due to the large search space. Particle swarm optimisation (PSO) is a powerful search technique, but there are some limitations on using the standard PSO for feature selection. This paper proposes a new PSO based feature selection approach, which introduces an external archive to store promising solutions obtained during the search process. The solutions in the archive serve as potential leaders (i.e. global best, gbest) to guide the swarm to search for an optimal feature subset with the lowest classification error rate and a smaller number of features. The proposed approach has two specific methods, PSOArR and PSOArRWS, where PSOArR randomly selects gbest from the archive and PSOArRWS uses the roulette wheel selection to select gbest considering both the classification error rate and also considering the number of selected features. Experiments on twelve benchmark datasets show that both PSOArR and PSOArRWS can successfully select a smaller number of features and achieve similar or better classification performance than using all features. PSOArR and PSOArRWS outperform a PSO based algorithm without using an archive and two traditional feature selection methods. The performance of PSOArR and PSOArRWS are similar to each other. }} @InProceedings{Sawczuk-da-Silva:2014:CEC, title = {A Graph-Based Particle Swarm Optimisation Approach to {QoS}-Aware Web Service Composition and Selection}, author = {Alexandre Sawczuk da Silva and Hui Ma and Mengjie Zhang}, pages = {3127--3134}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO)}, abstract = { Web services are network-accessible modules that perform specific tasks and can be integrated into Web service compositions to accomplish more complex objectives. Due to the fast-growing number of Web services and the well-defined nature of their interfaces, the field of automated Web service composition is quickly expanding. The use of Particle Swarm Optimisation composition techniques that take Quality of Service (QoS) properties into account is well-established in the field. However, the commonly utilised approach is to optimise a preselected Web service composition workflow, which requires domain expertise and prior knowledge and thus may lead to the loss of better solutions that require different workflow configurations. This paper presents a graph-based PSO technique which simultaneously determines an optimal workflow and near-optimal Web services to be included in the composition based on their QoS properties, as well as a greedy-based PSO technique which follows the commonly utilised approach. The comparison of the two techniques shows that despite requiring more execution time, the graph-based approach provides equivalent or better solutions than the greedy-based approach, depending on the workflow preselected by the greedy-based PSO. These results demonstrate that under certain circumstances, the graph-based approach is capable of producing solutions whose fitness surpasses that of the solutions obtained by employing the greedy-based approach. }} @InProceedings{Hardhienata:2014:CEC, title = {Task Allocation Under Communication Constraints Using Motivated Particle Swarm Optimization}, author = {Medria Hardhienata and Valery Ugrinovskii and Kathryn Merrick}, pages = {3135--3142}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO)}, abstract = { This paper considers task allocation problems where a group of agents must discover and allocate themselves to tasks. Task allocation is particularly difficult when agents can only exchange information over a limited communication range and when the agents are initialised from a single departure point. To address these constraints, we present a novel approach that incorporates computational models of motivation into a guaranteed convergence particle swarm optimisation algorithm. We introduce an incentive function and three motive profiles to guaranteed convergence particle swarm optimisation. Our new algorithm is compared to existing approaches with and without motivation under conditions of limited communication. It is tested in the case where the agents are initialised from a single point and random points. Results show that our approach increases the number of tasks discovered by a group of agents under these conditions. Furthermore, it significantly outperforms benchmark PSO algorithms in the number of tasks discovered and allocated when the agents are initialised from a single point. }} @InProceedings{McNabb:2014:CEC, title = {Serial {PSO} Results are Irrelevant in a Multi-Core Parallel World}, author = {Andrew McNabb and Kevin Seppi}, pages = {3143--3150}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Particle swarm optimisation (PSO), Parallel and distributed algorithms}, abstract = { From multi-core processors to parallel GPUs to computing clusters, computing resources are increasingly parallel. These parallel resources are being used to address increasingly challenging applications. This presents an opportunity to design optimisation algorithms that use parallel processors efficiently. In spite of the intuitively parallel nature of Particle Swarm Optimisation (PSO), many PSO variants are not evaluated from a parallel perspective and introduce extra communication and bottlenecks that are inefficient in a parallel environment. We argue that the standard practice of evaluating a PSO variant by reporting function values with respect to the number of function evaluations is inadequate for evaluating PSO in a parallel environment. Evaluating the parallel performance of a PSO variant instead requires reporting function values with respect to the number of iterations to show how the algorithm scales with the number of processors, along with an implementation-independent description of task interactions and communication. Furthermore, it is important to acknowledge the dependence of performance on specific properties of the objective function and computational resources. We discuss parallel evaluation of PSO, and we review approaches for increasing concurrency and for reducing communication which should be considered when discussing the scalability of a PSO variant. This discussion is essential both for designers who are defending the performance of an algorithm and for practitioners who are determining how to apply PSO for a given objective function and parallel environment. }} % Special Session: FrE4-3 Dynamic Multi-Objective Optimisation @InProceedings{Helbig:2014:CEC, title = {Heterogeneous Dynamic Vector Evaluated Particle Swarm Optimisation for Dynamic Multi-Objective Optimisation}, author = {Marde Helbig and Andries Engelbrecht}, pages = {3151--3159}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {PSO, Dynamic Multi-objective Optimisation}, abstract = { Optimisation problems with more than one objective, where at least one objective changes over time, are called dynamic multi-objective optimisation problems (DMOOPs). Since at least two objectives are in conflict with one another, a single solution does not exist, and therefore the goal of a dynamic multiobjective optimisation algorithm (DMOA) is to track the set of optimal trade-off solutions over time. One of the major issues when solving optimisation problems, is balancing exploration and exploitation during the search process. This paper investigates the performance of the dynamic vector evaluated particle swarm optimisation (DVEPSO) algorithm using heterogeneous PSOs (HPSOs), where each particle has a different behaviour. The goal of the study is to determine whether the use of heterogeneous particle swarm optimisation (HPSO) algorithms will improve the performance of DVEPSO by incorporating particles with exploration and exploitation behaviour in a single particle swarm optimisation (PSO) algorithm. The results indicate that using HPSOs improves the performance of DVEPSO, especially for type I and type III DMOOPs. }} @InProceedings{Liu:2014:CECk, title = {An Adaptive Diversity Introduction Method for Dynamic Evolutionary Multiobjective Optimization}, author = {Min Liu and Jinhua Zheng and Junnian Wang and Yuzhen Liu and Lei Jiang}, pages = {3160--3167}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic Multi-objective Optimisation}, abstract = { This paper investigates how to use diversity introduction methods to enhance the dynamic evolutionary multiobjective optimisation algorithms in dealing with dynamic multiobjective optimisation problems (DMOPs). Although diversity introduction method is easy used to response to the dynamic change, current diversity introduction methods still have a difficulty in identifying the correct proportion of diversity introduction. To overcome this difficulty, this paper proposes an adaptive diversity introduction (ADI) method. Specifically, the proportion of diversity introduction can be dynamically adjusted rather than being hand designed and fixed in advance. In addition, an adaptive relocation operator is designed to adapt the evolving individuals to the new environmental condition. The effectiveness of the ADI method is validated against various diversity introduction methods upon five DMOPs test problems. The simulation results show that the proposed ADI has better robustness and total performance than other diversity introduction methods. }} @InProceedings{Azzouz:2014:CEC, title = {A Multiple Reference Point-Based Evolutionary Algorithm for Dynamic Multi-Objective Optimization with Undetectable Changes}, author = {Radhia Azzouz and Slim Bechikh and Lamjed Ben Said}, pages = {3168--3175}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic Multi-objective Optimisation}, abstract = { Dynamic multi-objective optimisation problems involve the simultaneous optimisation of several competing objectives where the objective functions and/or constraints may change over time. Evolutionary algorithms have been considered as popular approaches to solve such problems. Despite the considerable number of studies reported in evolutionary optimisation in dynamic environments, most of them are restricted to the single objective case. Moreover, the majority of dynamic multi-objective optimisation algorithms are based on the use of some techniques to detect or predict changes which is sometimes difficult or even impossible. In this work, we address the problem of dynamic multi-objective optimisation with undetectable changes. To achieve this task, we propose a new algorithm called Multiple Reference Point-based Multi-Objective Evolutionary Algorithm (MRP-MOEA) which does not need to detect changes. Our algorithm uses a new reference point-based dominance relation ensuring the guidance of the search towards the Pareto optimal front. The performance of our proposed method is assessed using various benchmark problems. Furthermore, the comparative experiments show that MRP-MOEA outperforms several dynamic multi-objective optimisation algorithms not only in tracking the Pareto front but also in maintaining diversity over time albeit the changes are undetectable. }} @InProceedings{Rakshit:2014:CEC, title = {Artificial Bee Colony Induced Multi-Objective Optimization in Presence of Noise}, author = {Pratyusha Rakshit and Amit Konar and Atulya Nagar}, pages = {3176--3183}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Dynamic and uncertain environments, Multi-objective evolutionary algorithms, Numerical optimisation}, abstract = { The paper aims at designing new strategies to extend traditional Non-dominated Sorting Bee Colony algorithm to proficiently obtain Pareto-optimal solutions in presence of noise on the fitness landscapes. The first strategy, referred to as adaptive selection of sample-size, is employed to balance the trade-off between accurate fitness estimate and computational complexity. The second strategy is concerned with determining statistical expectation, instead of conventional averaging of fitness-samples as the measure of fitness of the trial solutions. The third strategy attempts to extend Goldberg's approach to examine possible placement of a slightly inferior solution in the optimal Pareto front using a more statistically viable comparator. Experiments undertaken to study the performance of the extended algorithm reveal that the extended algorithm outperforms its competitors with respect to four performance metrics, when examined on a test-suite of 23 standard benchmarks with additive noise of three statistical distributions. }} @InProceedings{Friedrich:2014:CEC, title = {A Cascaded Evolutionary Multi-Objective Optimization for Solving the Unbiased Universal Electric Motor Family Problem}, author = {Timo Friedrich and Stefan Menzel}, pages = {3184--3191}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Engineering applications, Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE)}, abstract = { For a successful business model the efficient development and design of a comprehensive product family plays a crucial part in many real world applications. A product family as it occurs, e.g., in the automotive domain consists of a product platform which covers the commonalities of product variants and the derived product variants. While product variants need to be fast and flexibly adjusted to market needs, from manufacturing and development point of view an underlying product platform with a large number of common parts is required to increase cost efficiency. For the design and evaluation of optimisation methods for product family development, in the present paper the universal electric motor (UEM) family problem is considered, as it provides a fair trade-off between complexity and computational costs compared to real world application scenarios in the automotive domain. Since especially solving this problem without usage of pre-knowledge comes with high computational costs, a cascaded evolutionary multiobjective optimisation based on NSGA-II with concatenation of product Pareto fronts is proposed in the present paper to efficiently reduce computational time. Besides providing sets of Pareto solutions to the unbiased UEM family problem the effects of considering solutions of prior platform optimisations as starting point for follow-up optimisations under changing requirements are evaluated. }} @InProceedings{Biswas:2014:CECa, title = {Evolutionary Multiobjective Optimization in Dynamic Environments: A Set of Novel Benchmark Functions}, author = {Subhodip Biswas and Swagatam Das and P. N. Suganthan and C. A. C Coello}, pages = {3192--3199}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary Multi-Objective Optimisation and Decision-Making, Multiobjective optimisation, Dynamic and uncertain environments}, abstract = { Time varying nature of the constraints, objectives and parameters characterise several practical optimisation problems. This fact leads us to the field of dynamic optimisation with Evolutionary Algorithms (EAs). In recent past, a significant amount of research has been devoted to single-objective dynamic optimisation problems. Very few researchers have, however, concentrated their efforts on the study of Dynamic multiobjective Optimisation Problems (DMOPs) where the dynamicity is attributed to multiple objectives of conflicting nature. Considering the lack of a somewhat diverse and challenging set of benchmark functions, in this article we discuss some techniques to design dynamic multiobjective problems. We propose some general techniques for introducing dynamicity in the Pareto Set and in the Pareto Front through shifting, shape variation, slope variation, phase variation, and several other types. We introduce nine benchmark functions, which have been derived from the benchmark suite used for the competition on bound-constrained and static multiobjective optimisation algorithms, held under the 2009 IEEE Congress on Evolutionary Computation (CEC). }} % Special Session: FrE4-4 Fireworks Algorithms for Optimisation @InProceedings{Zhang:2014:CECk, title = {A Hybrid Biogeography-Based Optimization and Fireworks Algorithm}, author = {Bei Zhang and Min-Xia Zhang and Yu-Jun Zheng}, pages = {3200--3206}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Fireworks Algorithms for Optimisation}, abstract = { The paper presents a hybrid biogeography-based optimisation (BBO) and fireworks algorithm (FWA) for global optimisation. The key idea is to introduce the migration operator of BBO to FWA, in order to enhance information sharing among the population, and thus improve solution diversity and avoid premature convergence. A migration probability is designed to integrate the migration of BBO and the normal explosion operator of FWA, which can not only reduce the computational burden, but also achieve a better balance between solution diversification and intensification. The Gaussian explosion of the enhanced FWA (EFWA) is reserved to keep the high exploration ability of the algorithm. Experimental results on selected benchmark functions show that the hybrid BBO\_FWA has a significantly performance improvement in comparison with both BBO and EFWA. }} @InProceedings{Liu:2014:CECl, title = {Analysis on Global Convergence and Time Complexity of Fireworks Algorithm}, author = {Jianhua Liu and Shaoqu Zheng and Ying Tan}, pages = {3207--3213}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Fireworks Algorithms for Optimisation, Convergence, scalability and complexity analysis, Evolutionary computation theory}, abstract = { Fireworks Algorithm (FWA) is a new proposed optimisation technique based on swarm intelligence. In FWA, the algorithm generates the explosion sparks and Gaussian mutation sparks by the explosion operator and Gaussian mutation operator to search the global optimum in the problem space. FWA has been applied in various fields of practical optimisation problems and gains great success. However, its convergence property has not been analysed since it has been provided. Same as other swarm intelligence (SI) algorithms, the optimisation process of FWA is able to be considered as a Markov process. In this paper, a Markov stochastic process on FWA has been defined, and is used to prove the global convergence of FWA while analysing its time complexity. In addition, the computation of the approximation region of expected convergence time of FWA has also been given. }} @InProceedings{Li:2014:CECq, title = {Adaptive Fireworks Algorithm}, author = {Junzhi Li and Shaoqiu Zheng and Ying Tan}, pages = {3214--3221}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Fireworks Algorithms for Optimisation}, abstract = { In this paper, firstly, the amplitude used in the Enhanced Fireworks Algorithm (EFWA) is analysed and its lack of adaptability is revealed, and then the adaptive amplitude method is proposed where amplitude is calculated according to the already evaluated fitness of the individuals adaptively. Finally, the Adaptive Fireworks Algorithm (AFWA) is proposed, replacing the amplitude operator in EFWA with the new adaptive amplitude. Some theoretical analyses are made to prove the adaptive explosion amplitude a promising method. Experiments on CEC13's 28 benchmark functions are also conducted in order to illustrate the performance and it turns out that the AFWA where adaptive amplitude is adopted outperforms significantly the EFWA and meanwhile the time consumed is not longer. Moreover, according to experimental results, AFWA performs better than the Standard Particle Swarm Optimisation (SPSO). }} @InProceedings{Zheng:2014:CECe, title = {Dynamic Search in Fireworks Algorithm}, author = {Shaoqiu Zheng and Andreas Janecek and Junzhi Li and Ying Tan}, pages = {3222--3229}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Fireworks Algorithms for Optimisation}, abstract = { We propose an improved version of the recently developed Enhanced Fireworks Algorithm (EFWA) based on an adaptive dynamic local search mechanism. In EFWA, the explosion amplitude (i.e., search area around the current location) of each firework is computed based on the quality of the firework's current location. This explosion amplitude is limited by a lower bound which decreases with the number of iterations in order to avoid the explosion amplitude to be [close to] zero, and in order to enhance global search abilities at the beginning and local search abilities towards the later phase of the algorithm. As the explosion amplitude in EFWA depends solely on the fireworks' fitness and the current number of iterations, this procedure does not allow for an adaptive optimisation process. To deal with these limitations, we propose the Dynamic Search Fireworks Algorithm (dynFWA) which uses a dynamic explosion amplitude for the firework at the currently best position. If the fitness of the best firework could be improved, the explosion amplitude will increase in order to speed up convergence. On the contrary, if the current position of the best firework could not be improved,the explosion amplitude will decrease in order to narrow the search area. In addition, we show that one of the EFWA operators can be removed in dynFWA without a loss in accuracy---this makes dynFWA computationally more efficient than EFWA. Experiments on 28 benchmark functions indicate that dynFWA is able to significantly outperform EFWA, and achieves better performance than the latest SPSO version SPSO2011. }} @InProceedings{Cheng:2014:CECb, title = {Maintaining Population Diversity in Brain Storm Optimization Algorithm}, author = {Shi Cheng and Yuhui Shi and Quande Qin and T. O. Ting and Ruibin Bai}, pages = {3230--3237}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Brain Storm Optimisation and Its Applications, Fireworks Algorithms for Optimisation, Swarm Intelligence for Real-world Engineering Optimisation}, abstract = { Swarm intelligence suffers the premature convergence, which happens partially due to the solutions getting clustered together, and not diverging again. The brain storm optimisation (BSO), which is a young and promising algorithm in swarm intelligence, is based on the collective behaviour of human being, that is, the brainstorming process. Premature convergence also happens in the BSO algorithm. The solutions get clustered after a few iterations, which indicate that the population diversity decreases quickly during the search. A definition of population diversity in BSO algorithm to measure the change of solutions' distribution is proposed in this paper. The algorithm's exploration and exploitation ability can be measured based on the change of population diversity. Two kinds of partial re-initialisation strategies are used to improve the population diversity in BSO algorithm. The experimental results show that the performance of the BSO is improved by these two strategies. }} @InProceedings{Yu:2014:CECh, title = {Fireworks Algorithm with Differential Mutation for Solving the {CEC 2014} Competition Problems}, author = {Chao Yu and Lingchen Kelley and Shaoqiu Zheng and Ying Tan}, pages = {3238--3245}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Single Objective Numerical Optimisation}, abstract = { The idea of fireworks algorithm (FWA) is inspired by the fireworks explosion in the sky at night. When a firework explodes, a shower of sparks appear around it. In this way, the adjacent area of the firework is searched. By controlling the amplitude of the explosion, the ability of local search for FWA is guaranteed. The way of fireworks algorithm searching the surrounding area can be further improved by differential mutation operator, forming an algorithm called FWA-DM. In this paper, the benchmark suite in the competition of congress of evolutionary computation (CEC) 2014 is used to test the performance of FWA-DM. }} % Session: FrE4-5 Real-World Applications III @InProceedings{Ivan:2014:CEC, title = {Evolutionary Algorithms Dynamics and Its Hidden Complex Network Structures}, author = {Zelinka Ivan and Lampinen Jouni and Senkerik Roman and Pluhacek Michal and Davendra Donald}, pages = {3246--3251}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Complex Networks and Evolutionary Computation, Evolutionary Computing with Deterministic Chaos}, abstract = { In this participation, we are continuing to show mutual intersection of two completely different areas of research: complex networks and evolutionary computation. Large-scale networks, exhibiting complex patterns of interaction amongst vertices exist in both nature and in man-made systems (i.e., communication networks, genetic pathways, ecological or economical networks, social networks, networks of various scientific collaboration etc. and are a part of our daily life. We are demonstrating that dynamics of evolutionary algorithms, that are based on Darwin theory of evolution and Mendel theory of genetic heritage, can be also visualised as a complex networks. Such network can be then analysed by means of classical tools of complex networks science. Results presented here are at the moment numerical demonstration rather than theoretical mathematical proofs. We open question whether evolutionary algorithms really create complex network structures and whether this knowledge can be successfully used like feedback for control of evolutionary dynamics and its improvement in order to increase the performance of evolutionary algorithms. }} @InProceedings{Suzuki:2014:CEC, title = {Knowledge Acquisition Issues for Intelligent Route Optimization by Evolutionary Computation}, author = {Masaki Suzuki and Setsuo Tsuruta and Rainer Knauf and Yoshitaka Sakurai}, pages = {3252--3257}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Evolutionary programming, Evolution strategies}, abstract = { The paper introduces a Knowledge Acquisition and Maintenance concept for a Case Based Approximation method to solve large scale Travelling Salesman Problems in a short time (around 3 seconds) with an error rate below 3\%. This method is based on the insight, that most solutions are very similar to solutions that have been created before. Thus, in many cases a solution can be derived from former solutions by (1) selecting a most similar TSP from a library of former TSP solutions, (2) removing the locations that are not part of the current TSP and (3) adding the missing locations of the current TSP by mutation, namely Nearest Insertion (NI). This way of creating solutions by Case Based Reasoning (CBR) avoids the computational costs to create new solutions from scratch. }} @InProceedings{Menezes:2014:CEC, title = {A Memetic Algorithm for the Prize Collecting Traveling Car Renter Problem}, author = {Matheus Menezes and Marco Goldbarg and Elizabeth Goldbarg}, pages = {3258--3265}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Memetic, multi-meme and hybrid algorithms, Heuristics, metaheuristics and hyper-heuristics}, abstract = { This paper introduces a new variant of the Travelling Car Renter Problem, named Prize-collecting Travelling Car Renter Problem. In this problem, a set of vertices, each associated with a bonus, and a set of vehicles are given. The bonus represents a degree of satisfaction to visit the vertex. The objective is to determine a cycle that visits some vertices collecting, at least, a predefined bonus, i.e. reaching a pre-specified satisfaction, and minimising the cost of the tour that can be traversed with different vehicles. A mathematical formulation is presented and implemented in a solver to produce results for sixty-four instances. A memetic algorithm is proposed and its performance is evaluated in comparison to the results obtained with the solver. }} @InProceedings{Wu:2014:CECg, title = {Network on Chip Optimization Based on Surrogate Model Assisted Evolutionary Algorithms}, author = {Mengyuan Wu and Ammar Karkar and Bo Liu and Alex Yakovlev and Georges Gielen}, pages = {3266--3271}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {EHW, Surrogate-assisted Global Optimisation Methods for Expensive Engineering Design}, abstract = { Network-on-Chip (NoC) design is attracting more and more attention nowadays, but there is a lack of design optimisation method due to the computationally very expensive simulations of NoC. To address this problem, an algorithm, called NoC design optimisation based on Gaussian process model assisted differential evolution (NDPAD), is presented. Using the surrogate model-aware evolutionary search (SMAS) framework with the tournament selection based constraint handling method, NDPAD can obtain satisfactory solutions using a limited number of expensive simulations. The evolutionary search strategies and training data selection methods are then investigated to handle integer design parameters in NoC design optimisation problems. Comparison shows that comparable or even better design solutions can be obtained compared to standard EAs, and much less computation effort is needed. }} @InProceedings{Liao:2014:CECa, title = {A Genetic Algorithm for the Minimum Latency Pickup and Delivery Problem}, author = {Xin-Lan Liao and Chih-Hung Chien and Chuan-Kang Ting}, pages = {3272--3279}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Intelligent Network Systems}, abstract = { The pickup and delivery problem combines vehicle routing and objects distribution to cope with logistic problems. While most research on PDP aims to minimise the transportation cost for the sake of service providers, this study proposes the minimum latency pickup and delivery problem (MLPDP) that seeks a low-latency route to transport commodities among nodes, where latency represents the sum of transportation time between consumers and the corresponding suppliers. The MLPDP is pertinent to time-sensitive services and logistics focusing on customer satisfaction. This study defines the latency of a customer as the average time elapsed aboard of goods received. The last-in-first-out loading method is employed to simulate real-world rear-loaded vehicles. This study further designs a genetic algorithm (GA) to resolve the MLPDP. In particular, we propose the edge aggregate crossover (EAC) and the reversely weighting technique to improve the performance of GA on the MLPDP. Experimental results show the effectiveness of the proposed GA. The results further indicate that EAC leads to significantly better performance than conventional crossover operators in solution quality and convergence speed on the MLPDP. }} @InProceedings{Weiszer:2014:CEC, title = {A Heuristic Approach to Greener Airport Ground Movement}, author = {Michal Weiszer and Jun Chen and Stefan Ravizza and Jason Atkin and Paul Stewart}, pages = {3280--3286}, booktitle = {Proceedings of the 2014 IEEE Congress on Evolutionary Computation}, year = {2014}, month = {6-11 July}, editor = {Carlos A. {Coello Coello}}, address = {Beijing, China}, ISBN = {0-7803-8515-2}, keywords = {Heuristic Methods for Multi-Component Optimisation Problems}, abstract = { Ever increasing air traffic, rising costs and tighter environmental targets create a pressure for efficient airport ground movement. Ground movement links other airport operations such as departure sequencing, arrival sequencing and gate/stand allocation and its operation can affect each of these. Previously, reducing taxi time was considered the main objective of the ground movement problem. However, this may conflict with efforts of airlines to minimise their fuel consumption as shorter taxi time may require higher speed and acceleration during taxiing. Therefore, in this paper a multi-objective multi-component optimisation problem is formulated which combines two components: scheduling and routing of aircraft and speed profile optimisation. To solve this problem an integrated solution method is adopted to more accurately investigate the trade-off between the total taxi time and fuel consumption. The new heuristic which is proposed here uses observations about the characteristics of the optimised speed profiles in order to greatly improve the speed of the graph-based routing and scheduling algorithm. Current results, using real airport data, confirm that this approach can find better solutions faster, making it very promising for application within on-line applications. }}