Construction of neural networks to approximate arbitrary continuous functions of one variable
Changwoon Choi, Jin Young Choi · Electronics Letters · 1992
A novel method is introduced for determining the connection weights of three layer neural networks which approximate any continuous functions of one variable. The derived formulas to calculate the weights explain the roles of hidden neurons and connection weights of three layer neural networks. It is demonstrated, by examples, that this method yields an easy and efficient construction of neural networks without needing training to approximate a function of one variable.