Neural Computation, Vol 9, 143-159, Copyright © 1997 by The MIT Press

LETTERS

Neural Networks for Functional Approximation and System Identification

HN Mhaskar and Nahmwoo Hahm

We construct generalized translation networks to approximate uniformly a class of nonlinear, continuous functionals defined on L^p ([-1,1]^s) for integer s 1, 1 p < infinity, or C([-1,1]^s). We obtain lower bounds on the possible order of approximation for such functionals in terms of any approximation process depending continuously on a given number of parameters. Our networks almost achieve this order of approximation in terms of the number of parameters (neurons) involved in the network. The training is simple and noniterative; in particular, we avoid any optimization such as that involved in the usual backpropagation.

This article has been cited by other articles:

				Y. Xia and M. S. Kamel A Measurement Fusion Method for Nonlinear System Identification Using a Cooperative Learning Algorithm Neural Comput., June 1, 2007; 19(6): 1589 - 1632. [Abstract] [Full Text] [PDF]