Abstract
Organic Computing (OC) aims at handling the growing complexity in technical systems by endowing them with life-like capabilities such as self-organisation, self-configuration, and self-adaptation. OC systems with these capabilities can tolerate disturbances and continue working properly while adapting their behaviour to the changes in their environment. In this context, the two-layer Observer/Controller architecture has been developed to determine the optimum set of parameters for an OC system that operates in a continuously changing environment. Layer 1 of this architecture, which is implemented using an eXtended Classifier System (XCS), allows for quick response to changes if a situation appears, which is close to a previously encountered situation. Thus, Layer 1 acts as a kind of memory. Layer 2 is triggered if the new situation is not covered by the population of the XCS on Layer 1. In that case, different parameter sets are evaluated using an optimisation algorithm on a simulation model of the real system. After that, the best parameter set found is given to the XCS on Layer 1 for further evaluation in the real world.
The contribution of this article is two-fold: Firstly, we present a rule combining mechanism for XCS that infers maximally general rules from the existing population to increase the on-line learning speed on Layer 1. Secondly, we present a new population-based optimisation algorithm for Layer 2, which can be used to find high quality solutions for OC systems that operate in continuously changing environments. Furthermore, we provide experimental results for both mechanisms and show that the proposed techniques improve both the learning rate and the solution quality.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Butz, M.V., Kovacs, T., Lanzi, P.L., Wilson, S.W.: Toward a theory of generalization and learning in XCS. IEEE Trans. Evol. Comput. 8(1), 28–46 (2004)
Cakar, E., Müller-Schloer, C.: Self-organising interaction patterns of homogeneous and heterogeneous multiagent populations. In: Proc. of the 3rd IEEE Int. Conf. on Self-Adaptive and Self-Organizing Systems, pp. 165–174 (2009)
Cakar, E., Tomforde, S., Müller-Schloer, C.: A role-based imitation algorithm for the optimisation in dynamic fitness landscapes. In: Swarm Intelligence Symposium, 2011. SIS 2011. IEEE, Paris, France (2011, accepted)
Eberhart, R.C., Kennedy, J.: Particle swarm optimization. In: Proc. of the 1995 IEEE Int. Conf. on Neural Networks, pp. 1942–1948 (1995)
Fredivianus, N., Prothmann, H., Schmeck, H.: XCS revisited: A novel discovery component for the eXtended Classifier System. In: Proceedings of the 8th Int. Conf. on Simulated Evolution And Learning. LNCS, vol. 6457, pp. 289–298. Springer, Berlin (2010)
Kirkpatrick, S., Gelatt, C.D., Jr., Vecchi, M.P.: Optimization by simulated annealing. Science 220, 671–680 (1983)
Müller-Schloer, C.: Organic computing—on the feasibility of controlled emergence. In: CODES+ISSS ’04: Proceedings of the International Conference on Hardware/Software Codesign and System Synthesis, pp. 2–5. IEEE Comput. Soc., Washington, DC (2004)
Nenortaite, J., Butleris, R.: Application of particle swarm optimization algorithm to decision making model incorporating cluster analysis. In: Human System Interactions, 2008 Conf., pp. 88–93 (2008)
North, M.J., Howe, T.R., Collier, N.T., Vos, J.R.: The repast symphony development environment. In: Proceedings of the Agent 2005 Conference on Generative Social Processes, Models and Mechanisms (2005)
Prothmann, H., Rochner, F., Tomforde, S., Branke, J., Müller-Schloer, C., Schmeck, H.: Organic control of traffic lights. In: Proceedings of the 5th International Conference on Autonomic and Trusted Computing (ATC-08). LNCS, vol. 5060, pp. 219–233. Springer, Berlin (2008)
Rejeb, L., Guessoum, Z., M’Hallah, R.: The exploration-exploitation dilemma for adaptive agents. In: Proceedings of the Fifth European Workshop on Adaptive Agents and Multi-Agent Systems (2005)
Storn, R., Price, K.: Differential evolution—a simple and efficient heuristic for global optimization over continuous spaces. J. Glob. Optim. 11(4), 341–359 (1997)
Ursem, R.K., Vadstrup, P.: Parameter identification of induction motors using differential evolution. In: The 2003 Congress on Evolutionary Computation, 2003. CEC ’03, vol. 2, pp. 790–796 (2003)
Watkins, C.: Learning from delayed reward. Ph.D. thesis (1989)
Wilson, S.W.: Generalization in the XCS classifier system. In: Genetic Programming 1998: Proceedings of the Third Annual Conference, pp. 665–674 (1998)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Basel AG
About this chapter
Cite this chapter
Cakar, E., Fredivianus, N., Hähner, J., Branke, J., Müller-Schloer, C., Schmeck, H. (2011). Aspects of Learning in OC Systems. In: Müller-Schloer, C., Schmeck, H., Ungerer, T. (eds) Organic Computing — A Paradigm Shift for Complex Systems. Autonomic Systems, vol 1. Springer, Basel. https://doi.org/10.1007/978-3-0348-0130-0_15
Download citation
DOI: https://doi.org/10.1007/978-3-0348-0130-0_15
Publisher Name: Springer, Basel
Print ISBN: 978-3-0348-0129-4
Online ISBN: 978-3-0348-0130-0
eBook Packages: Computer ScienceComputer Science (R0)