论文信息 - A new approach to fuzzy LCSs in two-dimensional continuous multistep environment with continuous vector actions

A new approach to fuzzy LCSs in two-dimensional continuous multistep environment with continuous vector actions

This paper introduces a new Fuzzy Learning Classifier System that can learn a continuous action vector field from two-dimensional continuous multistep environments. Results where this approach is applied with success to two simple two-dimensional problems are presented.

[1] Andrea Bonarini,et al. Evolutionary Learning of Fuzzy rules: competition and cooperation , 1996 .

[2] Manuel Valenzuela-Rendón,et al. The Fuzzy Classifier System: A Classifier System for Continuously Varying Variables , 1991, ICGA.

[3] Stewart W. Wilson. Three Architectures for Continuous Action , 2005, IWLCS.

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

[5] Manuel Valenzuela-Rendón,et al. Reinforcement learning in the fuzzy classifier system , 1998 .

[6] Alexandre Parodi,et al. A New Approach to Fuzzy Classifier Systems , 1993, ICGA.