Self-Adaptation in Learning Classifier Systems

The use and potential benefits of self-adaptive mutation operators are well-known within evolutionary computing. In this paper we begin by examining the use of self-adaptive mutation in Learning Classifier Systems. We implement the operator in the simple ZCS classifier and examine its performance in two maze environments. It is shown that, although no significant increase in performance is seen over results presented in the literature using a fixed rate of mutation, the operator adapts to an appropriate rate regardless of the initial range. The same concept is then applied to the learning rate parameter, but results show that a modification must be made to produce stable/effective controllers. Results from a fully self-adaptive system are also presented, with marked benefits found in a non-stationary environment. We then apply self-adaptation to the more complex XCS classifier system with similar overall results.