Design of Function Approximation Algorithm Based on RBF Neural Network

Abstract

Function approximation is an important part of function theory, and its role in numerical computation is crucial. The application of a neural network in function approximation is an innovative means of developing function approximation. However, most of the current researches compare the function approximation of various networks in depth. The purpose of this study is to investigate ways to examine and analyze function approximation based on a neural network, and describe a radial neural network. This study addresses the problem of function approximation algorithm design. Because this is based on an artificial neural network, a typical neural network is taken as an example, and the related concepts and algorithms are described in detail. The radial basis function neural network (RBFNN) for function approximation is designed and analyzed using experimental simulation. According to
the results of RBFNN and BPNN simulation experiments, the errors of RBFNN are close to 0, while the errors of BPNN have large oscillations around 0. From the results and analysis, it can be concluded that the approximation effect of RBFNN is better than that of BPNN, providing an important reference value for research on function approximation.

Keywords: function approximation, artificial neural network, radial neural network, BP neural network

Cite As

L. Hou, " Design of Function Approximation Algorithm Based on RBF Neural Network", Engineering Intelligent Systems, vol. 33 no. 2, pp. 155-167, 2025.