A modified RBF network with application to system identification

Proposes a modified RBF network in which the regression weights are used to replace the constant weights in the output layer. It is shown that the modified RBF network can reduce the number of hidden units significantly. Moreover, the authors develop a computationally efficient algorithm, known as the EM algorithm, to estimate the parameters of the regression weights. A salient feature of this algorithm is that it decouples a complicated multiparameter optimization problem into L separate small-scale optimization problems, where L is the number of hidden units. The superior performance the modified RBF network over the standard RBF network is illustrated by means of a system identification example.