Abstract
Learning Classifier Systems (LCSs) (Holland, 1976; Booker, Goldberg, & Holland, 1989) are rule-based evolutionary learning systems. A basic LCS consists of (1) a set of rules, that is, a population of classifiers, (2) a rule evaluation mechanism, which usually is realized by adapted reinforcement learning (RL) (Kaelbling, Littman, & Moore, 1996; Sutton & Barto, 1998) techniques, and (3) a rule evolution mechanism, which is usually implemented by a genetic algorithm (GA) (Holland, 1975). The classifier population codes the current knowledge of the LCS. The evaluation mechanism estimates and propagates rule utility. Based on the estimated utilities, the evolutionary mechanism generates offspring classifiers and deletes less useful classifiers.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer
About this chapter
Cite this chapter
Butz, M.V. (2006). Simple Learning Classifier Systems. In: Rule-Based Evolutionary Online Learning Systems. Studies in Fuzziness and Soft Computing, vol 191. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-31231-5_3
Download citation
DOI: https://doi.org/10.1007/3-540-31231-5_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-25379-2
Online ISBN: 978-3-540-31231-4
eBook Packages: EngineeringEngineering (R0)