Martin V. Butz, David E. Goldberg, & Wolfgang Stolzmann:
Investigating Generalization in the Anticipatory Classifier System
Recently, a genetic algorithm (GA) was introduced to the Anticipatory Classifier System (ACS) which surmounted the occasional problem of over-specialization of rules. This paper investigates the resulting generalization capabilities further by monitoring the performance of the ACS in the highly challanging multiplexer task in detail. Moreover, by comparing the ACS to the XCS classifier system in this task it is shown that the ACS generates accurate, maximally general rules and its population converges to those rules. Besides the observed ability of latent learning and the formation of an internal environmental representation, this ability of generalization adds a new advantage to the ACS in comparison with similar approaches.
In: Proceedings of the Sixth International Conference on Parallel Problem Solving from Nature. Berlin: Springer-Verlag.
generated: 24 February 2000; last update: 24 February 2000 / WST