Advertisement

Evolutionary Intelligence

, Volume 8, Issue 4, pp 185–201 | Cite as

An on-line Pittsburgh LCS for the Three-Cornered Coevolution Framework

  • Syahaneim MarzukhiEmail author
  • Will N. Browne
  • Mengjie Zhang
Research Paper
  • 119 Downloads

Abstract

The Three-Cornered Coevolution Framework describes a method that is capable of addressing classification tasks through coevolution (coadaptive evolution) where three different agents (i.e. a generation agent and two classification agents) learn and adapt to the changes of the problems without human involvement. Here, artificial problems can be generated in concert with classification agents in order to provide insight into their relationships. Previous work on the Two-Cornered Coevolution Framework provided foundation for implementing the system that was able to set-up the problem’s difficulty appropriately while triggering the coevolutionary process. However, the triggering process was set manually without utilising the third agent as proposed in the original framework to perform this task. Previous work on the Three-Cornered Coevolution introduced the third agent (a new classification agent) to trigger the coevolutionary process within the system, where its functionality and effect on the system requires investigation. This paper details the implementation for this case; two classification agents that use different styles of learning techniques (e.g. supervised versus reinforcement learning techniques) is adapted in the classification agents to learn the various classification problems. Ultimately, Learning Classifier System (LCS) is chosen to be implemented in the participating agents. LCS has several potential characteristics, such as interpretability, generalization capability and variations in representation, that are suitable for the system. Experiments show that the Pittsburgh-style LCS with the adaptation of Tabu Search technique in S capable to autonomously adjust the problem’s difficulty and generate a wide range of problems for classification. The adaptation of A-PLUS to an ‘on-line’ system is successful implemented. Further, the classification agents (i.e. R and I) are able to solve the classification tasks where the classification performance are varied. The Three-Cornered Coevolution Framework offers a great potential for autonomous learning and provides useful insight into coevolution learning over the standard studies of pattern recognition. The system is capable of autonomously generating various problems, learning and providing insight into each learning system’s ability by determining the problem domains where they perform relatively well. This is in contrast to humans having to determine the problem domains.

Keywords

Learning classifier systems Classification Coevolution Generation agent Classification agent 

References

  1. 1.
    Bacardit J, Krasnogor N (2009) Performance and efficiency of memetic Pittsburgh learning classifier systems. Evol Comput 17(3):307–342CrossRefGoogle Scholar
  2. 2.
    Bäck T, Fogel DB, Michalewicz Z (2000) Evolutionary computation 1: basic algorithms and operations. Institute of Physics Publishing, Bristol and PhiladelphiaGoogle Scholar
  3. 3.
    Bäck T, Fogel DB, Michalewicz Z (2000) Evolutionary computation 2: advanced algorithms and operations. Institute of Physics Publishing, Bristol and PhiladelphiaGoogle Scholar
  4. 4.
    Bernadó-Mansilla E, Garrell-Guiu JM (2003) Accuracy-based learning classifier systems: models, analysis and applications to classification tasks. Evol Comput 11(3):209–238CrossRefGoogle Scholar
  5. 5.
    Bishop CM (2006) Pattern recognition and machine learning, Natural Computing Series. Springer, BerlinGoogle Scholar
  6. 6.
    Butz MV (2006) Rule-based evolutionary online learning systems: a principal approach to LCS analysis and design. Springer, BerlinGoogle Scholar
  7. 7.
    Chen VC (2004) Evaluation of Bayes, ICA, PCA and SVM methods for classification. Technical Report RTO-MP-SET-080, Radar Division, US Naval Research Laboratory, 4555 Overlook Avenue, S.W. Washington DC 20375, USAGoogle Scholar
  8. 8.
    Duda RO, Hart PE, Stork DG (2001) Pattern classification, 2nd edn. Wiley, New YorkzbMATHGoogle Scholar
  9. 9.
    Engelbrecht AP (2005) Computational intelligence: an introduction, 2nd edn. Wiley, New YorkGoogle Scholar
  10. 10.
    Holland JH, Reitman JS (1978) Cognitive systems based on adaptive algorithms. In: Pattern directed inference systems. Academic Press, New York, pp 313–329Google Scholar
  11. 11.
    Jain AK, Duin RPW, Mao J (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal Mach Intell 22(1):4–35CrossRefGoogle Scholar
  12. 12.
    Jennings NR, Sycara KP, Wooldridge M (1998) A roadmap of agent research and development. Auton Agents Multi-Agent Syst 1(1):7–38CrossRefGoogle Scholar
  13. 13.
    Jong KAD, Spears WM, Gordon DF (1993) Using genetic algorithms for concept learning. Mach Learn 13:161–188Google Scholar
  14. 14.
    Kharbat F, Bull L, Odeh M (2005) Revisiting genetic selection in the XCS learning classifier system. In: IEEE congress on evolutionary computation, vol 3, pp 2061–2068Google Scholar
  15. 15.
    Kovacs T (2004) Rule fitness and pathology in learning classifier systems. Evol Comput 12(1):99–135CrossRefGoogle Scholar
  16. 16.
    Kukenys I, Browne WN, Zhang M (2011) Transparent, online image pattern classification using a learning classier system. In: European event on evolutionary computation in image analysis and signal processing (EvoApplications’11). Springer, BerlinGoogle Scholar
  17. 17.
    Maes P (1995) Artificial life meets entertainment: lifelike autonomous agents. Commun ACM 38(11):108–114CrossRefGoogle Scholar
  18. 18.
    Marzukhi S, Browne WN, Zhang M (2012) Two-cornered learning classifier systems for pattern generation and classification. In: Genetic and evolutionary computation conference (GECCO’12). ACM, pp 895–902Google Scholar
  19. 19.
    Marzukhi S, Browne WN, Zhang M (2013) Adaptive artificial datasets through learning classifier systems for classification tasks. In: International workshop in learning classifier systems (IWLCS’13). ACM, pp 1243–1250Google Scholar
  20. 20.
    Marzukhi S, Browne WN, Zhang M (2013) Adaptive artificial datasets through learning classifier systems for classification tasks. Evol Intell 6(2):93–107Google Scholar
  21. 21.
    Marzukhi S, Browne WN, Zhang M (2014) Three-cornered coevolution learning classifier systems for classification tasks. In: Genetic and evolutionary computation conference (GECCO’14). ACM, pp 549–556Google Scholar
  22. 22.
    Mohri M, Rostamizadeh A, Talwalkar A (2012) Foundations of machine learning. MIT Press, Cambridge, MAzbMATHGoogle Scholar
  23. 23.
    Neshatian K (2004) Strength or accuracy: credit assignment in learning classifier system, Ph.D. thesis. Bristol UniversityGoogle Scholar
  24. 24.
    Orriols-Puig A, Bernadó-Mansilla E (2005) The class imbalance problem in UCS classifier system: a preliminary study. In: International workshop learning classifier system (IWLCS’05), pp 161–180Google Scholar
  25. 25.
    Orriols-Puig A, Bernadó-Mansilla E (2006) A further look at UCS classifier system. In: Genetic and evolutionary computation conference (GECCO’06)Google Scholar
  26. 26.
    Orriols-Puig A, Bernadó-Mansilla E (2007) Revisiting UCS: description, fitness sharing, and comparison with XCS. In: International workshop learning classifier system (IWLCS’07), pp 96–116Google Scholar
  27. 27.
    Orriols-Puig A, Bernadó-Mansilla E (2009) Evolutionary rule-based systems for imbalanced data sets. Soft Comput 13(3):213–225CrossRefGoogle Scholar
  28. 28.
    Schürmann J (1996) Pattern classification: a unified view of statistical and neural approaches. Wiley, New YorkGoogle Scholar
  29. 29.
    Shafi K (2008) An online and adaptive signature-based approach for intrusion detection using learning classifier systems, Ph.D. thesis. School of Information Technology and Electrical Engineering, University of New South Wales, Australian Defence Force AcademyGoogle Scholar
  30. 30.
    Stacey A (2004) An investigation of techniques for improving the performance of the Pittsburgh LCS. Technical Report UWELCSG04-005, Department Computer Science, University of BathGoogle Scholar
  31. 31.
    Troc M, Unold O (2010) Self-adaptation of parameters in a learning classifier system ensemble machine. Appl Math Comput Sci 20(1):157–174zbMATHGoogle Scholar
  32. 32.
    Unold O (2010) Self-adaptive learning classifier system. J Circuits Syst Comput 19(1):275–296CrossRefGoogle Scholar
  33. 33.
    Wilson SW (1995) Classifier fitness based on accuracy. Evol Comput 3(2):149–175CrossRefGoogle Scholar
  34. 34.
    Wilson SW (2009) Coevolution of pattern generators and recognizers. In: International workshop learning classifier system, IWLCS 2009. ACM, New YorkGoogle Scholar
  35. 35.
    Yang J, Xu H, Jia P (2012) Effective search for Pittsburgh learning classifier systems via estimation of distribution algorithms. Inf Sci 198:100–117CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Syahaneim Marzukhi
    • 1
    Email author
  • Will N. Browne
    • 2
  • Mengjie Zhang
    • 2
  1. 1.Faculty of Science and Defence Technology, Computer Science DepartmentNational Defence University MalaysiaKuala LumpurMalaysia
  2. 2.Evolutionary Computation Research Group, School of Engineering and Computer ScienceVictoria University of WellingtonWellingtonNew Zealand

Personalised recommendations