This book constitutes the thoroughly refereed joint post-proceedings of 3 consecutive International Workshops on Learning Classifier Systems that took place in Chicago, IL, USA in July 2003, in Seattle, WA, USA in June 2004, and in Washington, DC, USA in June 2005 - all hosted by the Genetic and Evolutionary Computation Conference, GECCO.

The 22 revised full papers presented were carefully reviewed and selected from the workshop contributions. The papers are organized in topical sections on knowledge representation, mechanisms, new directions, as well as application-oriented research and tools. The topics range from theoretical analysis of mechanisms to practical consideration for successful application of such techniques to everday datamining tasks.

The 5th International Workshop on Learning Classi?er Systems (IWLCS2002) was held September 7-8, 2002, in Granada, Spain, during the 7th International Conference on Parallel Problem Solving from Nature (PPSN VII). We have included in this volume revised and extended versions of the papers presented at the workshop. In the ?rst paper, Browne introduces a new model of learning classi?er system, iLCS, and tests it on the Wisconsin Breast Cancer classi?cation problem. Dixon et al. present an algorithm for reducing the solutions evolved by the classi?er system XCS, so as to produce a small set of readily understandable rules. Enee and Barbaroux take a close look at Pittsburgh-style classi?er systems, focusing on the multi-agent problem known as El-farol. Holmes and Bilker investigate the effect that various types of missing data have on the classi?cation performance of learning classi?er systems. The two papers by Kovacs deal with an important theoretical issue in learning classi?er systems: the use of accuracy-based ?tness as opposed to the more traditional strength-based ?tness. In the ?rst paper, Kovacs introduces a strength-based version of XCS, called SB-XCS. The original XCS and the new SB-XCS are compared in the second paper, where - vacs discusses the different classes of solutions that XCS and SB-XCS tend to evolve.

This book constitutes the thoroughly refereed post-proceedings of the 4th International Workshop on Learning Classifier Systems, IWLCS 2001, held in San Francisco, CA, USA, in July 2001.The 12 revised full papers presented together with a special paper on a formal description of ACS have gone through two rounds of reviewing and improvement. The first part of the book is devoted to theoretical issues of learning classifier systems including the influence of exploration strategy, self-adaptive classifier systems, and the use of classifier systems for social simulation. The second part is devoted to applications in various fields such as data mining, stock trading, and power distributionn networks.

Learning classi er systems are rule-based systems that exploit evolutionary c- putation and reinforcement learning to solve di cult problems. They were - troduced in 1978 by John H. Holland, the father of genetic algorithms, and since then they have been applied to domains as diverse as autonomous robotics, trading agents, and data mining. At the Second International Workshop on Learning Classi er Systems (IWLCS 99), held July 13, 1999, in Orlando, Florida, active researchers reported on the then current state of learning classi er system research and highlighted some of the most promising research directions. The most interesting contri- tions to the meeting are included in the book Learning Classi er Systems: From Foundations to Applications, published as LNAI 1813 by Springer-Verlag. The following year, the Third International Workshop on Learning Classi er Systems (IWLCS 2000), held September 15{16 in Paris, gave participants the opportunity to discuss further advances in learning classi er systems. We have included in this volume revised and extended versions of thirteen of the papers presented at the workshop.

Learning Classifier Systems (LCS) are a machine learning paradigm introduced by John Holland in 1976. They are rule-based systems in which learning is viewed as a process of ongoing adaptation to a partially unknown environment through genetic algorithms and temporal difference learning. This book provides a unique survey of the current state of the art of LCS and highlights some of the most promising research directions. The first part presents various views of leading people on what learning classifier systems are. The second part is devoted to advanced topics of current interest, including alternative representations, methods for evaluating rule utility, and extensions to existing classifier system models. The final part is dedicated to promising applications in areas like data mining, medical data analysis, economic trading agents, aircraft maneuvering, and autonomous robotics. An appendix comprising 467 entries provides a comprehensive LCS bibliography.