Abstract: In order to remedy the inherent weaknesses of the back-propagation (13P) neural-network model and its learning algorithm,a multi-input I_agucrrcorthogonal-polynomial feed-forward neural network (MILOPNN) was constructed, which is based on the theory of polynomial interpolation and approximation. Then, a new kind of weights-and-structurcdctcrmination(WASD) algorithm was proposed to determine the optimal weights and structure of the MILOPNN ctuickly and automatically. Computer simulation and experiment results further substantiate the efficacy of the WASI)algorithm,as well as the relatively good abilities of approximation and denoising of the MILOPNN model equipped with the WASD algorithm.

Key words: Multi-input, Neural network, Laguerre-orthogonal-polynomial, Weights and structure determination, Optimal structure