Discovery of backpropagation learning rules using genetic programming

The backpropagation learning rule is a widespread computational method for training multilayer networks. Unfortunately, backpropagation suffers from several problems. In this paper, we have used Genetic Programming (GP) to overcome some of these problems and to discover new supervised learning algorithms. A set of such learning algorithms has been compared with the Standard BackPropagation (SBP) learning algorithm on different problems and has been shown to provide better performances. This study indicates that there exist many supervised learning algorithms better than, but similar to, SBP and that GP can be used to discover them.