INTEGRATED FEATURE AND ARCHITECTURE SELECTION

In this paper, we present an integrated approach to feature and archit ecture selection for single hidden layer-feedforward neural networks t rained via backpropagation. In our approach, we adopt a statistical mo del building perspective in which we analyze neural networks within a nonlinear regression framework, The algorithm presented in this paper employs a likelihood-ratio test statistic as a model selection criteri on, This criterion is used in a sequential procedure aimed at selectin g the best neural network given an initial architecture as determined by heuristic rules, Application results for an object recognition prob lem demonstrate the selection algorithm's effectiveness in identifying reduced neural networks with equivalent prediction accuracy.