Learning Classifier Systems

Albert Orriols (for those who do not know him, a brilliant learning classifier systems researcher) defended his thesis today. The outcome: Excellent Cum Laude. Albert Orriols Ph.D. defense started at 11am at Enginyeria i Arquitectura La Salle in Barcelona. The thesis panel was presided by Prof. David E. Goldberg, and formed by members Prof. Francisco Herrera, myself, Dr. Martin Butz, and secretary Prof. Xavier Vilasís). I must say that it has been a great pleasure to read his remarkable thesis and great contributions to the Learning Classifier System field. I hope he will make it available soon, and encourage you to take a look at it. ...

The 11th edition of the International Workshop on Learning Classifier System 2008 is hot. So far a lot of idea exchange and interesting discussions. So far Gilles Enee, myself, Natalio Krasnogor, Albert Oriols, Thyago Duque covering map problems, encoding language and model building, TSP and metaheuristics, learning association rules, and multi class labeling. Had to rush back to the next one :D

by Xavier Llorà and Kumara Sastry (2006, accepted). Proceedings of the ACM Genetic and Evolutionary Computation Conference (GECCO 2006), pp. 1513–1520, ACM press. Also as IlliGAL TR No 2006001. Link to the PDF. Abstract Over the last ten years XCS has become a de facto standard for Michigan-style learning classifier systems (LCS). Since the initial CS-1 work conceived by Holland, classifiers (rules) have widely used a ternary condition alphabet {0,1,#} for binary input problems. Most of the freely available implementations of this ternary alphabet in XCS rely on character-based encodings—easy to implement, not memory efficient, and expensive to compute. Profiling of freely available XCS implementations shows that most of their execution time is spent determining whether a rule is match or not, posing a serious thread to XCS scalability. In the last decade, multimedia and scientific applications have pushed CPU manufactures to include native support for vector instruction sets. This paper presents how to implement efficient condition encoding and fast rule matching strategies using vector instructions. The paper elaborates on Altivec (PowerPC G4, G5) and SSE2 (Intel P4/Xeon and AMD Opteron) instruction sets producing speedups of XCS matching process beyond ninety times. Moreover, such a vectorized matching code will allow to easily scale beyond tens of thousands of conditions in a reasonable time. The proposed fast matching scheme also fits in any other LCS other than XCS. ...

by Xavier Llorà, Kumara Sastry, and David E. Goldberg (2006). Proceedings of the Congress on Evolutionary Computation, 1, 596—603. Also as IlliGAL TR No 2005019. Link to the PDF. Abstract This paper presents an analysis of how maximally general and accurate rules can be evolved in a Pittsburgh-style classifier system. In order to be able to perform such an analysis we introduce a simple bare-bones Pittsburgh-style classifier systems—the compact classifier system (CCS)—based on estimation of distribution algorithms. Using a common rule encoding schemes of Pittsburgh-style classifier systems, CCS mantains a dynamic set of probability vectors that compactly describe a rule set. The compact genetic algorithm is used to evolve each of the initially perturbated probability vectors. Results show how CCS is able to evolve in a compact, simple, and elegant manner rule sets composed by maximally general and accurate rules. The initial theoretical analysis and results also show that traditional encoding schemes used by Pittsburgh-style classifiers add an extra facet of diffiiculty. Such a bias plays a central role on the overall performance and scalability of CCS and other Pittsburgh-style systems using such encoding schemes. ...