Initialization by Selection for Wavelet Network Training

Yacine Oussar (1) , Gérard Dreyfus (1)

1 SIGMA - Laboratoire Signaux, Modèles et Apprentissage Statistique

We present an original initialization procedure for the parameters of feedforward wavelet networks, prior to training by gradient-based techniques. It takes advantage of wavelet frames stemming from the discrete wavelet transform, 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 demonstrate the effectiveness of the proposed method.

Mots clés: Wavelet networks; Training; Initializing parameters; Nonlinear static modeling

Domaines: Apprentissage [cs.LG]

Yacine Oussar, Gérard Dreyfus. Initialization by Selection for Wavelet Network Training. Neurocomputing, 2000, 34 (1-4), pp.131-143. ⟨hal-00922196⟩