Introducing a Genetic Generalization Pressure to the Anticipatory Classifier System - Part 1: Theoretical approach

The Anticipatory Classifier System (ACS) is a learning classifier system that is based on the cognitive mechanism of anticipatory behavioral control. Besides the common reward learning, the ACS is able to learn latently (i.e. to learn without getting any reward) which is not possible with reinforcement learning techniques. Furthermore, it forms a complete internal representation of the environment and thus is able to use cognitive processes such as reasoning and planning. Latest research observed that the ACS is not generating accurate, maximally general rules reliably (i.e. rules which are accurate and also as general as possible), but it is sometimes generating over-specialized rules. This paper shows how a genetic algorithm can be used to overcome this present pressure of over-specialization in the ACS mechanism with a genetic generalization pressure. The ACS works then hybrid which learns latently, forms a cognitive map, and evolves accurate, maximally general rules.

[1]  Martin V. Butz,et al.  First Cognitive Capabilities in the Anticipatory Classifier System , 2000 .

[2]  Martin V. Butz,et al.  How XCS evolves accurate classifiers , 2001 .

[3]  Richard S. Sutton,et al.  Reinforcement learning architectures for animats , 1991 .

[4]  Pier Luca Lanzi,et al.  An Analysis of Generalization in the XCS Classifier System , 1999, Evolutionary Computation.

[5]  Stewart W. Wilson Generalization in the XCS Classifier System , 1998 .

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

[7]  Martin V. Butz Biasing Exploration in an Anticipatory Learning Classifier System , 2001, IWLCS.

[8]  David E. Goldberg,et al.  Introducing a Genetic Generalization Pressure to the Anticipatory Classiier System Part 2: Performance Analysis Introducing a Genetic Generalization Pressure to the Anticipatory Classiier System Part 2: Performance Analysis , 2007 .

[9]  Martin V. Butz,et al.  Latent Learning and Action Planning in Robots with Anticipatory Classifier Systems , 1999, Learning Classifier Systems.

[10]  D.E. Goldberg,et al.  Classifier Systems and Genetic Algorithms , 1989, Artif. Intell..

[11]  Martin V. Butz,et al.  An Algorithmic Description of ACS2 , 2001, International Workshop on Learning Classifier Systems.

[12]  er SystemsTim KovacsOctober Evolving Optimal Populations with XCS Classi , 1996 .

[13]  Kalyanmoy Deb,et al.  A Comparative Analysis of Selection Schemes Used in Genetic Algorithms , 1990, FOGA.

[14]  Martin V. Butz,et al.  Anticipations Control Behavior: Animal Behavior in an Anticipatory Learning Classifier System , 2002, Adapt. Behav..

[15]  M. Pelikán,et al.  Analyzing the evolutionary pressures in XCS , 2001 .

[16]  Martin V. Butz,et al.  Probability-Enhanced Predictions in the Anticipatory Classifier System , 2000, IWLCS.

[17]  Wolfgang Stolzmann Antizipative classifier systems , 1997 .

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

[19]  Martin V. Butz,et al.  Introducing a Genetic Generalization Pressure to the Anticipatory Classifier System Part 2: Performa , 2000 .

[20]  Stewart W. Wilson ZCS: A Zeroth Level Classifier System , 1994, Evolutionary Computation.

[21]  D. Goldberg,et al.  New Challenges for an Anticipatory Classiier System: Hard Problems and Possible Solutions New Challenges for an Anticipatory Classiier System: Hard Problems and Possible Solutions , 2007 .

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