Advertisement

Robustness and Evolution in an Adaptive System Application on Classification Task

  • J. Biondi
Conference paper

Abstract

In this paper, we proposed an approach to a single-step Classifier System, in which the useful population is built by progressively specializing classifiers. It has been applied to a classification task in a medical domain. To permit the system to explore alternatives without making decisions earlier in learning stages, all the classifiers that might be selected are triggered and receive the resulting reward corresponding to their action. The payoff function involves the classifier’s performance, its specificity and the system’s performance (its robustness). Genetic operators are activated with a probability which depends on the system’s robustness. During the test stages, no further learning takes place and the system’s performance is measured by the percentage of correct classification made on the second set of examples. When the measure of performance is the highest, the population is stabilized and contains the correct classifiers (the payoff function and genetic operators have no more effect on classifiers). This approach achieves convergency more quickly and makes it possible to have a final accurate population without over-specializing.

Keywords

Payoff Function Classifier System Crossover Operator Genetic Operator Testing Stage 
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

(a)Book

  1. [1]
    Goldberg D E, ‘Genetic Algorithms in search, Optimisation, and Machine Learning’, Addison Wesley, Reading, MA, 1989.Google Scholar
  2. [2]
    Davis L, ‘Handbook of Genetic Algorithms’, Van Nostrand Reinhold, New York, 1991.Google Scholar

(b)Article

  1. [3]
    Biondi J, ‘MAGE, un Modèle d’Apprentissage GEnéral’, thèse d’Etat, Univerité de Nice-Sophia Antipolis, 1990.Google Scholar
  2. [4]
    Bonelli P and Parodi A, ‘An efficient Classifier System and its Experimental Comparison with two Representative learning methods on three medical domains’, Proceedings of the 4th ICGA, Morgan Kaufmann Publishers, San Mateo, California, 1991.Google Scholar
  3. [5]
    Booker L B, ‘Improving the performance of genetic algorithms in classifier systems.’, Proceedings of the first ICGA, Hillsdale, New jersey, Lawrence Erlbaum Associates, 1985.Google Scholar
  4. [6]
    De Jong K A and Spears W M, ‘Learning Concept Classification Rules Using Genetic Algorithms’, Proceedings of 12th IJCAI, Sydney, Australia, 1991.Google Scholar
  5. [7]
    Grefenstette J J, ‘A system for Learning Control Strategies with Genetic Algorithms’, Proceedings of the 3rd ICGA, Morgan Kaufmann Publishers, Inc., San Mateo, California, 1989.Google Scholar
  6. [8]
    Grefenstette J J, ‘Lamarkian Learning in Multiagent Environments’, Proceedings of the 4th ICGA, Morgan Kaufmann Publishers, California 1991.Google Scholar
  7. [9]
    Holland J H, ‘Adaptation in Natural and Artificial Sytem’, The University of Michigan Press, Ann Arbour, 1975.Google Scholar
  8. [10]
    Holland J H, ‘Escaping Brittleness: The Possibilities of General-Purpose Learning Algorithms Applied to Parallel Rule-Based Systems’, in Machine Learning II, R.S. Michalski, J. G. Carbonell and T.M. Mitchell (ed.), 1986.Google Scholar
  9. [11]
    Kelly J D and Davis L, ‘A Hybrid Genetic Algorithm for Classification’, Proceeding of 12th IJCAI91, Sydney, Australia, 1991.Google Scholar
  10. [12]
    Rescorla R A and Wagner A R, ‘Variations in the effectiveness of reinforcement ans nonreinforcement’, in Classical Conditioning II: Current Research and Theory, Ed. A.H. Black and W.F. Prokasy, pp. 64–99, New York, Appleton-century-Crofts, 1972.Google Scholar
  11. [13]
    Shu L and Schaeffer J, ‘VCS: Variable Classifier Systems’, Proceedings of the 3rd ICGA, Morgan Kaufmann Publishers, 1989.Google Scholar
  12. [14]
    Shu L and Schaeffer J, ‘HCS: Adding Hierarchies to Classifier Systems’, Proceedings of the 4th ICGA, 1991Google Scholar
  13. [15]
    Schuurmans D and Schaeffer J, ‘Representational difficulties with Classifier Systems’, Proceedings of the 3rd ICGA, Morgan Kaufmann Publishers, 1989.Google Scholar
  14. [16]
    Sedbrook T A, Wright H and Wright R, ‘Application of a Genetic Classifier for Patient Triage’, Proceedings of the 4th ICGA, 1991.Google Scholar
  15. [17]
    Valenzuela Rendon M, ‘The Fuzzy Classifier System: A Classifier System for Continuously Varying Variables’, 4th ICGA, 1991.Google Scholar
  16. [18]
    Wilson S W, ‘Classifier Systems and the Animat Problem’, in Machine Learning Journal 2, Kluwer Academic Publishers, Boston, 1987.Google Scholar
  17. [19]
    Zhou H H and Grefenstette J J., ‘Learning by Analogy in Genetic Classifier Systems’, Proceedings of the 3rd ICGA, Morgan Kaufmann Publishers, Inc., San Mateo, California, 1989.Google Scholar

Copyright information

© Springer-Verlag/Wien 1993

Authors and Affiliations

  • J. Biondi
    • 1
  1. 1.University of Nice-Sophia Antipolis, I3S-CnrsValbonneFrance

Personalised recommendations