Initialization by selection for wavelet network training

We present an original initialization procedure for the parameters of feedf orward wavelet networks, prior to training by gradient-based techniques. It takes advantage of wavelet frames stemming from the discrete wavelet trans form, and uses a selection method to determine a set of best wavelets whose centers and dilation parameters are used as initial values for subsequent training. Results obtained for the modeling of two simulated processes are compared to those obtained with a heuristic initialization procedure, and t he effectiveness of the proposed method is demonstrated. (C) 2000 Elsevier Science B.V. All rights reserved.