On the Role of Regularization Parameters in Fitness Functions for Evolutionary Designed Artificial N

A main criterion for the accuracy of solutions of Artiicial Neural Networks (ANNs) for classii-cation tasks is the architecture. In order to nd problem-adapted topologies of ANNs, we adopted the evolutionary approach to ANN design by employing a Genetic Algorithm (GA) to evolve ANNs which are represented using a direct encoding method. The role of tness functions used by the GA is investigated, especially, the impact of a tness function expressing both, the learning error and a toplogy-dependent regularization term, is studied. A parallel system { the netGEN system { which has been implemented by the authors is generating problem{adapted Feed{Forward ANNs being trained by Error{Back{Propagation. Empirical results on a real world problem taken from the PROBEN1 ANN benchmark suite are presented.