wuerzburg.de

The XCS classifier system has been successfully applied to various problem domains including datamining, boolean classifications, and function approximation. In all these ap-plications just two classifiers were reproduced in a match or action set, given a time-recency threshold was met in the set. In this paper, we investigate the effect of selecting more than two classifiers for reproduction in XCSF. We either in-crease the number of selected classifiers or select a number of classifiers relative to the current match set size. In the functions investigated, both approaches showed a highly sig-nificant increase in initial learning speed. Also, in less chal-lenging approximation tasks, the final accuracy reached is not affected by the approach. However, in harder functions, learning may stall due to over-reproductions of inaccurate, ill-estimated classifiers. Thus, we propose an adaptive off-spring size rate that may depend on the current reliability of classifier parameter estimates. First results with a fixed offspring set size decrement show promising results. Future work is needed to speed-up XCS’s learning progress and ad-just its learning speed to the perceived problem difficulty