Rule-Based Evolutionary Online Learning Systems

1. Front Matter

   Pages I-XXI

   PDF

2. Introduction

   • Martin V. Butz

   Pages 1-7

3. Prerequisites

   • Martin V. Butz

   Pages 9-30

4. Simple Learning Classifier Systems

   • Martin V. Butz

   Pages 31-50

5. The XCS Classifier System

   • Martin V. Butz

   Pages 51-64

6. How XCS Works: Ensuring Effective Evolutionary Pressures

   • Martin V. Butz

   Pages 65-90

7. When XCS Works: Towards Computational Complexity

   • Martin V. Butz

   Pages 91-122

8. Effective XCS Search: Building Block Processing

   • Martin V. Butz

   Pages 123-146

9. XCS in Binary Classification Problems

   • Martin V. Butz

   Pages 147-156

10. XCS in Multi-Valued Problems

    • Martin V. Butz

    Pages 157-179

11. XCS in Reinforcement Learning Problems

    • Martin V. Butz

    Pages 181-195

12. Facetwise LCS Design

    • Martin V. Butz

    Pages 197-206

13. Towards Cognitive Learning Classifier Systems

    • Martin V. Butz

    Pages 207-217

14. Summary and Conclusions

    • Martin V. Butz

    Pages 219-225

15. Back Matter

    Pages 227-266

    PDF

Rule-basedevolutionaryonlinelearningsystems,oftenreferredtoasMichig- style learning classi?er systems (LCSs), were proposed nearly thirty years ago (Holland, 1976; Holland, 1977) originally calling them cognitive systems. LCSs combine the strength of reinforcement learning with the generali- tion capabilities of genetic algorithms promising a ?exible, online general- ing, solely reinforcement dependent learning system. However, despite several initial successful applications of LCSs and their interesting relations with a- mal learning and cognition, understanding of the systems remained somewhat obscured. Questions concerning learning complexity or convergence remained unanswered. Performance in di?erent problem types, problem structures, c- ceptspaces,andhypothesisspacesstayednearlyunpredictable. Thisbookhas the following three major objectives: (1) to establish a facetwise theory - proachforLCSsthatpromotessystemanalysis,understanding,anddesign;(2) to analyze, evaluate, and enhance the XCS classi?er system (Wilson, 1995) by the means of the facetwise approach establishing a fundamental XCS learning theory; (3) to identify both the major advantages of an LCS-based learning approach as well as the most promising potential application areas. Achieving these three objectives leads to a rigorous understanding of LCS functioning that enables the successful application of LCSs to diverse problem types and problem domains. The quantitative analysis of XCS shows that the inter- tive, evolutionary-based online learning mechanism works machine learning competitively yielding a low-order polynomial learning complexity. Moreover, the facetwise analysis approach facilitates the successful design of more - vanced LCSs including Holland’s originally envisioned cognitive systems. Martin V.

• Artificial Intelligence / Machine Learning
• Cognitive Science
• Genetic Algorithms
• Genetics-Based Machine Learning
• Learning Classifier Systems
• Online Learning
• Reinforcement Learning
• algorithm
• algorithms
• calculus
• classification
• complexity
• data mining
• learning

• Department of Cognitive Psychology, University of Würzburg, Würzburg, Germany

  Martin V. Butz