Advertisement

State of XCS Classifier System Research

  • Stewart W. Wilson
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1813)

Abstract

XCS is a new kind of learning classifier system that differs from the traditional kind primarily in its definition of classifier fitness and its relation to contemporary reinforcement learning. Advantages of XCS include improved performance and an ability to form accurate maximal generalizations. This paper reviews recent research on XCS with respect to representation, internal state, predictive modeling, noise, and underlying theory and technique. A notation for environmental regularities is introduced.

Keywords

Learning Classifier System Internal Exploration Minimum Prediction Error Bucket Brigade History Window 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Booker, L.: Intelligent behavior as an adaptation to the task environment. Ph.D. Dissertation (Computer and Communication Sciences). The University of Michigan (1982).Google Scholar
  2. 2.
    Cliff, D. and Ross, S.: Adding temporary memory to ZCS. Adaptive Behavior 3(2) (1994) 101–150.CrossRefGoogle Scholar
  3. 3.
    Goldberg, D. E., Korb, B., and Deb, K.: Messy genetic algorithms: motivation, analysis, and first results. Complex Systems 3 (1989) 493–530.zbMATHMathSciNetGoogle Scholar
  4. 4.
    Holland, J. H.: Escaping brittleness: the possibilities of general-purpose learning algorithms applied to parallel rule-based systems. In: R. S. Michalski, J. G. Carbonell & T. M. Mitchell (Eds.), Machine Learning, An Artificial Intelligence Approach. Volume II. Los Altos, California: Morgan Kaufmann (1986).Google Scholar
  5. 5.
    Holland, J. H.: Concerning the emergence of tag-mediated lookahead in classifier systems. Physica D, 41 (1990) 188–201.CrossRefGoogle Scholar
  6. 6.
    Kovacs, T.: Evolving Optimal Populations with XCS Classifier Systems. MSc. Dissertation, Univ. of Birmingham, UK (1996).Google Scholar
  7. 7.
    Kovacs, T.: XCS classifier system reliably evolves accurate, complete, and minimal representations for Boolean functions. In: Roy, Chawdhry and Pant (Eds), Soft Computing in Engineering Design and Manufacturing, Springer-Verlag London (1997) 59–68.Google Scholar
  8. 8.
    Kovacs, T.: Deletion schemes for classifier systems. In: Banzhaf, W., et al, (eds.), Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-99). Morgan Kaufmann: San Francisco, CA (1999) 329–336.Google Scholar
  9. 9.
    Kovacs, T.: Strength or accuracy? A comparison of two approaches to fitness calculation in learning classifier systems. This volume.Google Scholar
  10. 10.
    Koza, J. R.: Genetic Programming. Cambridge, MA: The MIT Press/Bradford Books (1992).zbMATHGoogle Scholar
  11. 11.
    Lanzi, P. L.: Adding memory to XCS. In: Proceedings of the IEEE Conference on Evolutionary Computation (ICEC98). IEEE Press (1998).Google Scholar
  12. 12.
    Lanzi, P. L.: An analysis of the memory mechanism of XCSM In J. Koza et al (eds.), Proceedings of the Third Annual Genetic Programming Conference, San Francisco, CA: Morgan Kaufmann (1998) 643–651.Google Scholar
  13. 13.
    Lanzi, P. L.: Extending the representation of classifier conditions, part I: from binary to messy coding. In: Banzhaf, W., et al, (eds.), Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-99). Morgan Kaufmann: San Francisco, CA (1999) 337–344.Google Scholar
  14. 14.
    Lanzi, P. L.: Reinforcement Learning with Classifier Systems. PhD Thesis, Politecnico di Milano (1999).Google Scholar
  15. 15.
    Lanzi, P. L.: An analysis of generalization in the XCS classifier system. Evolutionary Computation, 7(2) (1999) 125–149.CrossRefGoogle Scholar
  16. 16.
    Lanzi, P. L. and Colombetti, M.: An extension to the XCS classifier system for stochastic environments. In: Banzhaf, W., et al, (eds.), Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-99). Morgan Kaufmann: San Francisco, CA (1999) 353–360.Google Scholar
  17. 17.
    Lanzi, P. L. and Perrucci, A.: Extending the representation of classifier conditions, part II: from messy coding to s-expressions. In: Banzhaf, W., et al, (eds.), Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-99). Morgan Kaufmann: San Francisco, CA (1999) 345–352.Google Scholar
  18. 18.
    Lanzi, P. L. and Wilson, S. W.: Toward optimal classifier system performance in non-Markov environments. Evolutionary Computation, to appear (2000).Google Scholar
  19. 19.
    Lin, L.-J.: Reinforcement Learning for Robots Using Neural Networks. Ph.D. Dissertation (School of Computer Science), Carnegie Mellon University (1993).Google Scholar
  20. 20.
    Riolo, R. L.: Lookahead planning and latent learning in a classifier system. In J.-A. Meyer and S. W. Wilson (eds.), From Animals to Animats: Proceedings of the First International Conference on Simulation of Adaptive Behavior. Cambridge, MA: MIT Press (1991) 316–326.Google Scholar
  21. 21.
    Robertson, G. G. and Riolo, R. L.: A tale of two classifier systems. Machine Learning, 3 (1988) 139–159.Google Scholar
  22. 22.
    Smith, R. E.: Default Hierarchy Formation and Memory Expoitation in Learning Classifier Systems. Ph. D. Dissertation, The University of Alabama, Tuscaloosa, Alabama (1991).Google Scholar
  23. 23.
    Stolzmann, W.: Anticipatory classifier systems. In: Proceedings of the Third Annual Genetic Programming Conference, J. Koza et al (eds.). San Francisco, CA: Morgan Kaufmann (1988) 658–664.Google Scholar
  24. 24.
    Sutton, R. S.: Reinforcement learning architectures for animats. In: J.-A. Meyer and S. W. Wilson (eds.), From Animals to Animats: Proceedings of the First International Conference on Simulation of Adaptive Behavior. Cambridge, MA: MIT Press (1991) 288–296.Google Scholar
  25. 25.
    Sutton, R. S. and Barto A. G.: Reinforcement Learning: An Introduction. Cambridge, MA: The MIT Press/Bradford Books (1998).Google Scholar
  26. 26.
    Teller, A.: The evolution of mental models. In: K. E. Kinnear, Jr. (ed.), Advances in Genetic Programming. Cambridge, MA: The MIT Press/Bradford Books (1994).Google Scholar
  27. 27.
    Watkins, C. J. C. H.: Learning from Delayed Rewards. Ph.D. Dissertation, Cambridge University (1989).Google Scholar
  28. 28.
    Wilson, S. W.: Hierarchical credit allocation in a classifier system. In: Proceedings of the Tenth International Joint Conference on Artificial Intelligence. Los Altos, CA: Morgan Kaufmann (1987) 217–220.Google Scholar
  29. 29.
    Wilson, S. W.: ZCS: a zeroth level classifier system. Evolutionary Computation 2(1) (1994) 1–18.CrossRefGoogle Scholar
  30. 30.
    Wilson, S. W.: Classifier fitness based on accuracy. Evolutionary Computation 3(2) (1995) 149–175.CrossRefGoogle Scholar
  31. 31.
    Wilson, S. W.: Generalization in the XCS classifier system. In Proceedings of the Third Annual Genetic Programming Conference, J. Koza et al (eds.). San Francisco, CA: Morgan Kaufmann (1998) 665–674.Google Scholar
  32. 32.
    Wilson, S. W.: Get real! XCS with continuous-valued inputs. This volume.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Stewart W. Wilson
    • 1
    • 2
  1. 1.The University of IllinoisUrbana-ChampaignUSA
  2. 2.Prediction DynamicsConcordUSA

Personalised recommendations