Robert Junges
ABSTRACT
Developing a valid agent-based simulation model is not always straight forward, but involves a lot of prototyping, testing and analyzing until the right low-level behavior is fully specified and calibrated. Our aim is to replace the try and error search of a modeler by adaptive agents which learn a behavior that then can serve as a source of inspiration for the modeler. In this contribution, we suggest to use genetic programming as the learning mechanism. For this aim we developed a genetic programming framework integrated into the visual agent-based modeling and simulation tool SeSAm, providing similar easy-to-use functionality.
- M. V. Butz, K. Sastry, and D. E. Goldberg. Strong, stable, and reliable fitness pressure in xcs due to tournament selection. Genetic Programming and Evolvable Machines, 6:53--77, March 2005. Google Scholar
Digital Library
- J. Decraene, Y. Y. Cheng, M. Low, S. Zhou, W. Cai, and C. S. Choo. Evolving agent-based simulations in the clouds. In Advanced Computational Intelligence (IWACI), 2010 Third International Workshop on, pages 244 --249, August 2010.Google ScholarCross Ref
- A. E. Eiben and J. E. Smith. Introduction to Evolutionary Computing. Natural Computing Series. Springer, 2003. Google ScholarDigital Library
- J. M. Epstein and R. L. Axtrell. Growing Artificial Societies: Social Science from the Bottom Up. MIT Press, 1996. Google ScholarDigital Library
- M. Fehler, F. Klügl, and F. Puppe. Learning for analysis and calibration in agent-based simulations. In Proc. of ALAMAS Workshop, Brussels, 2006, 2006.Google Scholar
- T. Francisco and G. M. Jorge dos Reis. Evolving predator and prey behaviours with co-evolution using genetic programming and decision trees. In Proceedings of the 2008 GECCO conference companion on Genetic and evolutionary computation, GECCO '08, pages 1893--1900, New York, NY, USA, 2008. ACM. Google ScholarDigital Library
- L. Hanna and J. Cagan. Evolutionary multi-agent systems: An adaptive and dynamic approach to optimization. Journal of Mechanical Design, 131(1):011010, 2009.Google ScholarCross Ref
- W. H. Hsu and S. M. Gustafson. Genetic programming and multi-agent layered learning by reinforcements. In In Genetic and Evolutionary Computation Conference, pages 764--771. Morgan Kaufmann, 2002. Google ScholarDigital Library
- R. Junges and F. Klügl. Evaluation of techniques for a learning-driven modeling methodology in multiagent simulation. In MATES, pages 185--196, 2010. Google ScholarDigital Library
- R. Junges and F. Klügl. Learning convergence and agent behavior interpretation for designing agent-based simulations. In Proceedings of the Eighth European Workshop on Multi-Agent Systems EUMAS 2010, 2010.Google Scholar
- F. Klügl. Agent-based simulation engineering. Habilitation Thesis, University of Würzburg, unpublished, available online via the homepage of the author, 2009.Google Scholar
- F. Klügl, R. Hatko, and M. V. Butz. Agent learning instead of behavior implementation for simulations - a case study using classifier systems. In MATES 2008: Proceedings of the 6th German Conference on Multiagent System Technologies, pages 111--122. Springer Berlin / Heidelberg, 2008. Google ScholarDigital Library
- F. Klügl, R. Herrler, and M. Fehler. Sesam: implementation of agent-based simulation using visual programming. In AAMAS '06: Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems, pages 1439--1440, New York, NY, USA, 2006. ACM. Google ScholarDigital Library
- J. R. Koza. Genetic Programming: On the Programming of Computers by Means of Natural Selection (Complex Adaptive Systems). The MIT Press, 1 edition, Dec. 1992. Google ScholarDigital Library
- M. R. Lee and E.-K. Kang. Learning enabled cooperative agent behavior in an evolutionary and competitive environment. Neural Computing & Applications, 15:124--135, 2006. Google ScholarDigital Library
- S. Mahadevan and J. Connell. Automatic programming of behavior-based robots using reinforcement learning. Artificial Intelligence, 55(2-3):311--365, 1992. Google ScholarDigital Library
- R. Neruda, S. Slusny, and P. Vidnerova. Performance comparison of relational reinforcement learning and rbf neural networks for small mobile robots. In FGCNS '08: Proceedings of the 2008 Second International Conference on Future Generation Communication and Networking Symposia, pages 29--32, Washington, DC, USA, 2008. IEEE Computer Society. Google ScholarDigital Library
- D. J. Nowak and G. B. Lamont. Autonomous agent behavior generation using multiobjective evolutionary optimization. In Proceedings of the 2008 GECCO conference companion on Genetic and evolutionary computation, GECCO '08, pages 1961--1968, New York, NY, USA, 2008. ACM. Google ScholarDigital Library
- A. Nowe, K. Verbeeck, and M. Peeters. Learning automata as a basis for multi agent reinforcement learning. pages 71--85, 2006. Google ScholarDigital Library
- C. Oechslein, A. Hörnlein, and F. Klügl. Evolutionary optimization of simulated multi-agent systems. In Proc. of MASHO ("Modelling Artificial Societies and Hybrid Organisations"), Workshop held at the ECAI2000, Berlin August 2000, 2000.Google Scholar
- S. Schulenburg and P. Ross. Strength and money: An lcs approach to increasing returns. In Advances in Learning Classifier Systems, pages 291--298. Springer Berlin / Heidelberg, 2001. Google Scholar
- R. S. Sutton and A. G. Barto. Reinforcement Learning: An Introduction. MIT Press, 1998. Google ScholarDigital Library
- G. Weiß. Adaptation and learning in multi-agent systems: Some remarks and a bibliography. In IJCAI '95: Proceedings of the Workshop on Adaption and Learning in Multi-Agent Systems, pages 1--21, London, UK, 1996. Springer-Verlag. Google ScholarDigital Library
Index Terms
- Evolution for modeling: a genetic programming framework for sesam
Recommendations
Modeling and Simulation for Mission Operations Work System Design
Work system analysis and design is complex and nondeterministic. In this paper we describe Brahms, a multiagent modeling and simulation environment for designing complex interactions in human-machine systems. Brahms was originally conceived as a ...
Learning agent models in SeSAm
AAMAS '13: Proceedings of the 2013 international conference on Autonomous agents and multi-agent systemsDesigning the agent model in a multiagent simulation is a challenging task due to the generative nature of such systems. In this contribution we present an extension to the multiagent simulation platform SeSAm, introducing a learning-based design ...
Optimizing Context Computation for Multiagent Simulations
WI-IAT '14: Proceedings of the 2014 IEEE/WIC/ACM International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT) - Volume 03The execution of a multiagent-based simulation (MABS) model necessitates a scheduler that synchronizes the agents execution and simulates the simultaneity of their behaviors. In the majority of MABS frameworks the scheduler activates the agents who ...
Comments