Resilient backpropagation for RBF networks

Many algorithms have been proposed in order to train Radial Basis Function (RBF) networks. In this paper, Resilient Backpropagation (RPROP) with a weight decay term is proposed to train RBF networks and used to differentiate surfaces of 3D object in range images and to classify eight different Machine Learning data set for classification purpose. We show the advantages of resilient backpropagation for the RBF network structure within this classification context. The network structure is a combination of supervised and unsupervised learning layers. Experimental results show that radial basis function network trained with resilient backpropagation can be successfully applied to differentiate of surfaces of 3D object in range images as well as to the classification of Machine learning problems.