Evolving Temporal Rules with the Delayed Action Classifier System — Analysis and New Results

Conference paper


The Delayed Action Classifier System (DACS) is a rule-based system which employs evolutionary algorithms to discover temporal rules for operation in environments with a rich temporal structure. In the DACS architecture, rules encode timing information about when the actions of rules should be taken. This paper offers a mathematical analysis which demonstrates the advantages of DACS compared with traditional chained classifier systems in terms of the likelihood of discovery of rule-bases that can accurately represent time. An extended version of DACS (called DACS2) is described and justified. DACS2 is experimentally evaluated. Areas for further development of the system are discussed, and possible applications of the system in the domain of manufacturing and design are suggested.


Classifier System Internal Clock Temporal Classifier Classifier Chain Fitness Sharing 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Moore-Ede MC, Sulzman FM, Fuller CA (1982) The clocks that time us. Harvard University Press, Cambridge, MAGoogle Scholar
  2. 2.
    Krebs JR, Kacelnik A (1984) Time horizons of foraging animals. In Timing and Time Perception, Annals of the New York Academy of Sciences, 423:278–291CrossRefGoogle Scholar
  3. 3.
    Gibbon J, Church RM, Meck WH (1984) Scalar timing in memory. In Timing and Time Perception, Annals of the New York Academy of Sciences, 423:52–77CrossRefGoogle Scholar
  4. 4.
    Carse B (1994) Learning anticipatory behaviour using a delayed action classifier system. In: Fogarty TC (ed.) Procs Evolutionary Computing: Selected Papers of the AISB Workshop, Springer-Verlag, Berlin Heidelberg, 210–223Google Scholar
  5. 5.
    Carse B, Fogarty TC (1994) Genetic search in the design of rule-based controllers with temporal and predictive behaviour. In: Parmee IC (ed.) Procs of Conference on Adaptive Computing in Engineering Design and Control, University of Plymouth, pp 56–60Google Scholar
  6. 6.
    Holland JH (1975) Adaptation in natural and artificial systems. University of Michigan Press, Ann Arbour, MIGoogle Scholar
  7. 7.
    Booker LB, Goldberg DE, Holland JH (1989) Classifier systems and genetic algorithms. Artificial Intelligence 40:235–282CrossRefGoogle Scholar
  8. 8.
    Holland JH, Holyoak KJ, Nisbett RE, Thagard PR (1987) Induction: processes of inference, learning, and discovery, MIT PressGoogle Scholar
  9. 9.
    Robertson GG, Riolo RL (1988) A tale of two classifier systems. In: Goldberg DE, Holland JH (eds.) Machine Learning 3:139–159, Kluwer Academic PublishersGoogle Scholar
  10. 10.
    Wilson SW, Goldberg DE (1989) A critical review of classifier systems. Procs of the 3rd International Conference on Genetic Algorithms, pp244–255Google Scholar
  11. 11.
    Shu L, Schaeffer J (1991) HCS: Adding hierarchies to classifier systems. In: Belew RK, Booker LB (eds.) Procs of the 4th International Conference on Genetic Algorithms, pp339–345Google Scholar
  12. 12.
    Tomlinson A, Bull L (1999) On Corporate classifier systems: improving the use of rule-linkage. In: Banzhaf W, Daida J, Eiben AE, Garzon MH, Honavar V, Jakiela M, Smith RE (eds.) GECCO-99: Procs of the Genetic and Evolutionary Computation Conference. Morgan Kaufmann, pp649–656Google Scholar
  13. 13.
    Cobb HG, Grefenstette JJ (1991) Learning the persistence of actions in reactive control rules. In: Procs of the 8th International Machine Learning Workshop, Morgan Kaufmann, pp 293–297Google Scholar
  14. 14.
    Barry A (2000) Specifying action persistence in XCS. In: GECCO-2000: Procs of the Genetic and Evolutionary Computation Conference, Morgan Kaufmann, San Francisco, pp 50–57Google Scholar
  15. 15.
    Wilson SW (1995) Classifier Fitness based on Accuracy. Evolutionary Computing 3(2):149–175CrossRefGoogle Scholar
  16. 16.
    Carse B, Fogarty TC, Munro A (1998) Artificial evolution of fuzzy rule bases which represent time: a temporal fuzzy classifier system. International Journal of Intelligent Systems, 13(10/11):905–927CrossRefGoogle Scholar
  17. 17.
    Stolzmann W (1998) Anticipatory classifier systems. In: Genetic Programming ′98, University of Wisconsin, Madison, Wisconsin. Morgan Kaufmann, pp658–664Google Scholar
  18. 18.
    Goldberg DE, Richardson J (1987) Genetic algorithms with sharing for multimodal function optimisation. In: Procs of the 2nd International Conference on Genetic Algorithms, pp41–49Google Scholar
  19. 19.
    Gan J, Warwick K (2000) A Variable radius niching technique for speciation in genetic algorithms. In GECCO-2000: Procs of the Genetic and Evolutionary Computation Conference, Morgan Kaufmann, San Francisco, pp 96–103Google Scholar

Copyright information

© Springer-Verlag London 2002

Authors and Affiliations

  1. 1.Intelligent Autonomous Systems Laboratory Faculty of ComputingEngineering and Mathematical Sciences University of the West of EnglandBristolEngland

Personalised recommendations