Supervised Learning Errors by Radial Basis Function Neural Networks and Regularization Networks

Abstract

There is a gap between the theoretical results of regularization theory and practical suitability of regularization-derived networks (RN). On the other hand, radial basis function networks (RBF) that can be seen as a special case of regularization networks, have a rich selection of learning algorithms. In this work we study a relationship between RN and RBF, and show that theoretical estimates for RN hold for a concrete RBF applied on real-world data.