Abstract
We propose in this paper a modification of one of the modern state-of-the-art genetic programming algorithms used for data-driven modeling, namely the Bi-objective Genetic Programming (BioGP). The original method is based on a concurrent minimization of both the training error and complexity of multiple candidate models encoded as Genetic Programming trees. Also, BioGP is empowered by a predator-prey co-evolutionary model where virtual predators are used to suppress solutions (preys) characterized by a poor trade-off error vs complexity. In this work, we incorporate in the original BioGP an adaptive mechanism that automatically tunes the mutation rate, based on a characterization of the current population (in terms of entropy) and on the information that can be extracted from it. We show through numerical experiments on two different datasets from the energy domain that the proposed method, named BioAGP (where “A” stands for “Adaptive”), performs better than the original BioGP, allowing the search to maintain a good diversity level in the population, without affecting the convergence rate.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
We should observe, however, that the main drawback of data-driven modeling methods is that they depend entirely on experimental data. Therefore, such methods can only be applied after the actual building is built and measurements are available.
- 2.
In the rest of the paper, we will consider problems with p = 1 output variable. Nonetheless multiple output variables can be approximated by multiple GP trees, one per variable. An extension of the analysis for p > 1 is also possible and will be considered in future studies.
- 3.
- 4.
We should remark that among all the Pareto-optimal solutions returned by BioAGP, we report here the ones characterized by the lowest training error, following the approach suggested in [11]. However other choices e.g. based on information criteria can also be made.
References
Behbahani, S., De Silva, C.W.: Mechatronic design evolution using bond graphs and hybrid genetic algorithm with genetic programming. IEEE/ASME Trans. Mechatron. 18(1), 190–199 (2013)
Bevilacqua, V., Cassano, F., Mininno, E., Iacca, G.: Optimizing feed-forward neural network topology by multi-objective evolutionary algorithms: a comparative study on biomedical datasets. In: Rossi, F., Mavelli, F., Stano, P., Caivano, D. (eds.) WIVACE 2015. CCIS, vol. 587, pp. 53–64. Springer, Heidelberg (2016). doi:10.1007/978-3-319-32695-5_5
Caraffini, F., Neri, F., Iacca, G., Mol, A.: Parallel memetic structures. Inf. Sci. 227, 60–82 (2013)
Caraffini, F., Neri, F., Passow, B.N., Iacca, G.: Re-sampled inheritance search: high performance despite the simplicity. Soft. Comput. 17(12), 2235–2256 (2013)
Coello, C.A.C.: Multi-objective evolutionary algorithms in real-world applications: some recent results and current challenges. In: Advances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences, pp. 3–18. Springer (2015)
Dasgupta, D., Michalewicz, Z.: Evolutionary Algorithms in Engineering Applications. Springer Science & Business Media (2013)
Deb, K.: Multi-objective Optimization Using Evolutionary Algorithms, vol. 16. Wiley, Chichester (2001)
Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans. Evol. Comput. 6(2), 182–197 (2002)
Ferreira, P., Ruano, A., Silva, S., Conceio, E.: Neural networks based predictive control for thermal comfort and energy savings in public buildings. Energy Build. 55, 238–251 (2012)
Garg, A., Tai, K.: Comparison of regression analysis, artificial neural network and genetic programming in handling the multicollinearity problem. In: International Conference on Modelling, Identification & Control (ICMIC), pp. 353–358. IEEE (2012)
Giri, B.K., Hakanen, J., Miettinen, K., Chakraborti, N.: Genetic programming through bi-objective genetic algorithms with a study of a simulated moving bed process involving multiple objectives. Appl. Soft Comput. 13(5), 2613–2623 (2013)
Heaton, J.: Programming Neural Networks with Encog 2 in Java (2010)
Iacca, G.: Distributed optimization in wireless sensor networks: an island-model framework. Soft. Comput. 17(12), 2257–2277 (2013)
Iacca, G., Caraffini, F., Neri, F.: Memory-saving memetic computing for path-following mobile robots. Appl. Soft Comput. 13(4), 2003–2016 (2013)
Iacca, G., Caraffini, F., Neri, F.: Multi-strategy coevolving aging particle optimization. Int. J. Neural Syst. 24(01), 1450008 (2014)
Iacca, G., Mininno, E.: Introducing Kimeme, a novel platform for multi-disciplinary multi-objective optimization. In: Rossi, F., Mavelli, F., Stano, P., Caivano, D. (eds.) WIVACE 2015. CCIS, vol. 587, pp. 40–52. Springer, Heidelberg (2016). doi:10.1007/978-3-319-32695-5_4
Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge (1992)
Lichman, M.: UCI Machine Learning Repository (2013). http://archive.ics.uci.edu/ml
Menolascina, F., Tommasi, S., Paradiso, A., Cortellino, M., Bevilacqua, V., Mastronardi, G.: Novel data mining techniques in aCGH based breast cancer subtypes profiling: the biological perspective. In: IEEE Symposium on Computational Intelligence and Bioinformatics and Computational Biology (CIBCB), pp. 9–16, April 2007
Menolascina, F., Bellomo, D., Maiwald, T., Bevilacqua, V., Ciminelli, C., Paradiso, A., Tommasi, S.: Developing optimal input design strategies in cancer systems biology with applications to microfluidic device engineering. BMC Bioinform. 10(12), 1 (2009)
Onwubolu, G.C., Babu, B.: New optimization techniques in engineering, 141 (2013). Springer
Parmee, I.C.: Evolutionary and Adaptive Computing in Engineering Design. Springer Science & Business Media, London (2012)
Rennard, J.P.: Handbook of research on nature-inspired computing for economics and management. IGI Global (2006)
Riedmiller, M., Braun, H.: RPROP-a fast adaptive learning algorithm. In: Proceedings of ISCIS VII, Universitat (1992)
Costa e Silva, M.A., Coelho, L.d.S., Lebensztajn, L.: Multiobjective biogeography-based optimization based on predator-prey approach. IEEE Trans. Magn. 48(2), 951–954 (2012)
Stadler, W.: Multicriteria Optimization in Engineering and in the Sciences, vol. 37. Springer Science & Business Media, New York (2013)
Stanley, K.O., Miikkulainen, R.: Evolving neural networks through augmenting topologies. Evol. Comput. 10(2), 99–127 (2002)
Tsanas, A., Xifara, A.: Accurate quantitative estimation of energy performance of residential buildings using statistical machine learning tools. Energy Build. 49, 560–567 (2012)
Crepinsek, M., Liu, S.H., Mernik, M.: Exploration and exploitation in evolutionary algorithms: a survey. ACM Comput. Surv. 45(3), 35:1–35:33 (2013)
Vrieze, S.I.: Model selection and psychological theory: a discussion of the differences between the Akaike Information Criterion (AIC) and the Bayesian information criterion (BIC). Psychol. Methods 17(2), 228 (2012)
Zamora-Martnez, F., Romeu, P., Botella-Rocamora, P., Pardo, J.: On-line learning of indoor temperature forecasting models towards energy efficiency. Energy Build. 83, 162–172 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Bevilacqua, V., Nuzzolese, N., Mininno, E., Iacca, G. (2016). Adaptive Bi-objective Genetic Programming for Data-Driven System Modeling. In: Huang, DS., Han, K., Hussain, A. (eds) Intelligent Computing Methodologies. ICIC 2016. Lecture Notes in Computer Science(), vol 9773. Springer, Cham. https://doi.org/10.1007/978-3-319-42297-8_24
Download citation
DOI: https://doi.org/10.1007/978-3-319-42297-8_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-42296-1
Online ISBN: 978-3-319-42297-8
eBook Packages: Computer ScienceComputer Science (R0)