Pier Luca Lanzi
ABSTRACT
We study how different prediction update algorithms influence the performance of XCSF. We consider three classical parameter estimation algorithms (NLMS, RLS, and Kalman filter) and four gain adaptation algorithms (K1, K2, IDBD, and IDD). The latter have been shown to perform comparably to the best algorithms (RLS and Kalman), but they have a lower complexity. We apply these algorithms to update classifier prediction in XCSF and compare the performances of the seven versions of XCSF on a set of real functions. Our results show that the best known algorithms still perform best: XCSF with RLS and XCSF with Kalman perform significantly better than the others. In contrast, when added to XCSF, gain adaptation algorithms perform comparably to NLMS, the simplest estimation algorithm, the same used in the original XCSF. Nevertheless, algorithms that perform similarly generalize differently. For instance: XCSF with Kalman filter evolves more compact solutions than XCSF with RLS and gain adaptation algorithms allow better generalization than NLMS.
- M. V. Butz, K. Sastry, and D. E. Goldberg. Strong, stable, and reliable fitness pressure in XCS due to tournament selection. Genetic Programming and Evolvable Machines, 6(1):53--77, 2005. Google Scholar
Digital Library
- M. V. Butz and S. W. Wilson. An algorithmic description of XCS. Journal of Soft Computing, 6(3-4):144--153, 2002.Google ScholarCross Ref
- S. A. Glantz and B. K. Slinker. Primer of Applied Regression & Analysis of Variance. McGraw Hill, 2001. second edition.Google Scholar
- G. C. Goodwin and K. S. Sin. Adaptive Filtering: Prediction and Control. Prentice-Hall information and system sciences series, Mar. 1984.Google Scholar
- M. E. Harmon and L. C. B. III. Multi-player residual advantage learning with general function. Technical report, Air Force Base Ohio: Wright Laboratory, 1996.Google Scholar
- S. Haykin. Adaptive Filter Theory. Prentice-Hall, 2001. 4th Edition. Google ScholarDigital Library
- P. L. Lanzi, D. Loiacono, S. W. Wilson, and D. E. Goldberg. Extending XCSF beyond linear approximation. In Genetic and Evolutionary Computation - GECCO-2005, pages 1859--1866, Washington DC, USA, 2005. ACM Press. Google ScholarDigital Library
- P. L. Lanzi, D. Loiacono, S. W. Wilson, and D. E. Goldberg. Generalization in the XCSF classifier system: Analysis, improvement, and extension. Technical Report 2005012, Illinois Genetic Algorithms Laboratory - University of Illinois at Urbana-Champaign, 2005.Google Scholar
- P. L. Lanzi, D. Loiacono, S. W. Wilson, and D. E. Goldberg. XCS with computed prediction for the learning of Boolean functions. In Proceedings of the IEEE Congress on Evolutionary Computation - CEC-2005, pages 588--595, Edinburgh, UK, Sept. 2005. IEEE.Google ScholarCross Ref
- P. L. Lanzi, D. Loiacono, S. W. Wilson, and D. E. Goldberg. Prediction update algorithms for XCSF: Rls, kalman filter, and gain adaptation. Technical Report 2006008, Illinois Genetic Algorithms Laboratory - University of Illinois at Urbana-Champaign, 2006.Google Scholar
- R. S. Sutton. Adapting bias by gradient descent: An incremental version of delta-bar-delta. In Proceedings of the Tenth National Conference on Artificial Intelligence, pages 171--176. MIT Press, 1992.Google Scholar
- R. S. Sutton. Gain adaptation beats least squares? In Proceedings of the Seventh Yale Workshop on Adaptive and Learning Systems, pages 161--166. Yale University, New Haven, CT, 1992.Google Scholar
- B. Widrow and M. E. Hoff. Adaptive Switching Circuits, chapter Neurocomputing: Foundation of Research, pages 126--134. The MIT Press, Cambridge, 1988. Google ScholarDigital Library
- S. W. Wilson. Classifiers that approximate functions. Journal of Natural Computating, 1(2-3):211--234, 2002. Google ScholarDigital Library
Index Terms
- Prediction update algorithms for XCSF: RLS, Kalman filter, and gain adaptation
Recommendations
Extending XCSF beyond linear approximation
GECCO '05: Proceedings of the 7th annual conference on Genetic and evolutionary computationXCSF is the extension of XCS in which classifier prediction is computed as a linear combination of classifier inputs and a weight vector associated to each classifier. XCSF can exploit classifiers' computable prediction to evolve accurate piecewise ...
Self-adaptive mutation in XCSF
GECCO '08: Proceedings of the 10th annual conference on Genetic and evolutionary computationRecent advances in XCS technology have shown that self-adaptive mutation can be highly useful to speed-up the evolutionary progress in XCS. Moreover, recent publications have shown that XCS can also be successfully applied to challenging real-valued ...
Intrusion detection with evolutionary learning classifier systems
Evolutionary Learning Classifier Systems (LCSs) combine reinforcement learning or supervised learning with effective genetics-based search techniques. Together these two mechanisms enable LCSs to evolve solutions to decision problems in the form of easy ...
Comments