Evolving Classifiers Ensembles with Heterogeneous Predictors

XCS with computed prediction, namely XCSF, extends XCS by replacing the classifier prediction with a parametrized prediction function. Although several types of prediction functions have been introduced, so far XCSF models are still limited to evolving classifiers with the same prediction function. In this paper, we introduce XCSF with heterogeneous predictors, XCSFHP, which allows the evolution of classifiers with different types of prediction function within the same population. We compared XCSFHP to XCSF on several problems. Our results suggest that XCSFHP generally performs as XCSF with the most appropriate prediction function for the given problem. In particular, XCSFHP seems able to evolve, in each problem subspace, the most adequate type of prediction function.

[1]  Daniele Loiacono,et al.  XCSF with Neural Prediction , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[2]  Stewart W. Wilson.  Mining Oblique Data with XCS , 2000, IWLCS.

[3]  Stewart W. Wilson.  Classifier Systems for Continuous Payoff Environments , 2004, GECCO.

[4]  Mark Jurik,et al.  Neurocomputing: Foundations of research , 1992 .

[5]  Riccardo Poli,et al.  Genetic and Evolutionary Computation – GECCO 2004 , 2004, Lecture Notes in Computer Science.

[6]  Daniele Loiacono,et al.  Extending XCSF beyond linear approximation , 2005, GECCO '05.

[7]  Stewart W. Wilson.  Classifiers that approximate functions , 2002, Natural Computing.

[8]  Larry Bull,et al.  On using ZCS in a Simulated Continuous Double-Auction Market , 1999, GECCO.

[9]  Stewart W. Wilson.  Classifier Fitness Based on Accuracy , 1995, Evolutionary Computation.

[10]  Andrew W. Moore,et al.  Generalization in Reinforcement Learning: Safely Approximating the Value Function , 1994, NIPS.

[11]  Matthew Studley,et al.  Learning Classifier System Ensembles With Rule-Sharing , 2007, IEEE Transactions on Evolutionary Computation.

[12]  Martin V. Butz,et al.  An algorithmic description of XCS , 2000, Soft Comput..

[13]  Martin V. Butz,et al.  An Algorithmic Description of XCS , 2000, IWLCS.

[14]  Xavier Llorà,et al.  Mixed Decision Trees: Minimizing Knowledge Representation Bias in LCS , 2004, GECCO.

[15]  Tim Kovacs,et al.  Proceedings of the Third International Workshop on Learning Classifier Systems (IWLCS-2000), in the Joint Workshops of SAB 2000 and PPSN 2000 , 2000 .

[16]  Daniele Loiacono,et al.  XCS with computed prediction in multistep environments , 2005, GECCO '05.

[17]  Larry Bull,et al.  On the use of rule-sharing in learning classifier system ensembles , 2005, 2005 IEEE Congress on Evolutionary Computation.

[18]  Daniele Loiacono,et al.  Evolving Neural Networks for Classifier Prediction with XCSF , 2006 .

[19]  Marco Colombetti,et al.  Robot Shaping: An Experiment in Behavior Engineering , 1997 .

[20]  Daniele Loiacono,et al.  Support vector regression for classifier prediction , 2007, GECCO '07.

[21]  Daniele Loiacono,et al.  Generalization in the XCSF Classifier System: Analysis, Improvement, and Extension , 2007, Evolutionary Computation.

[22]  Daniele Loiacono,et al.  XCS with computed prediction in continuous multistep environments , 2005, 2005 IEEE Congress on Evolutionary Computation.

[23]  Tim Kovacs,et al.  Advances in Learning Classifier Systems , 2001, Lecture Notes in Computer Science.

[24]  Bernard Widrow,et al.  Adaptive switching circuits , 1988 .

[25]  Martin V. Butz.  Kernel-based, ellipsoidal conditions in the real-valued XCS classifier system , 2005, GECCO '05.

[26]  Daniele Loiacono,et al.  XCS with computed prediction for the learning of Boolean functions , 2005, 2005 IEEE Congress on Evolutionary Computation.

[27]  Anthony J. Bagnall,et al.  Using an Adaptive Agent to Bid in a Simplified Model of the UK Market in Electricity , 1999, Annual Conference on Genetic and Evolutionary Computation.