FUNDAMENTALNAYA I PRIKLADNAYA MATEMATIKA

(FUNDAMENTAL AND APPLIED MATHEMATICS)

2009, VOLUME 15, NUMBER 3, PAGES 9-21

Neural network approximation of several variable functions

D. V. Alexeev

Abstract

View as HTML     View as gif image

The main result of the work is as follows: in the Chebyshev--Hermite weighted integral metric, it is possible to approximate any function of sufficiently general form by neural network. The approximating net consists of two layers, where the first uses any predefined sigmoid function of activation and second uses linear-threshold function. The Chebyshev--Hermit weight is chosen because it lets one to imitate receptors distribution for example in an eye of a human or some mammal.

Location: http://mech.math.msu.su/~fpm/eng/k09/k093/k09303h.htm
Last modified: January 13, 2010