Title :
Research on the Non-Linear Function Fitting of RBF Neural Network
Author :
Liu Jin-Yue ; Zhu Bao-Ling
Author_Institution :
Comput. & Inf. Technol. Coll., Northeast Pet. Univ., Daqing, China
Abstract :
By the simulation instance, this paper carries out a comparative research of the function approximation ability of BP network and RBF network, and analyzes the fitting accuracy and time efficiency of these two artificial neural networks when they are used to accomplish nonlinear function fitting under the specified parameters. The results show that the function approximation ability of BP network is superior to BR network in many ways.
Keywords :
backpropagation; function approximation; nonlinear functions; radial basis function networks; BP network; RBF neural network; artificial neural networks; function approximation ability; nonlinear function fitting; simulation instance; Biological neural networks; Function approximation; Least squares approximations; Radial basis function networks; Training; BP neural network; RBF neural network; function approximation;
Conference_Titel :
Computational and Information Sciences (ICCIS), 2013 Fifth International Conference on
Conference_Location :
Shiyang
DOI :
10.1109/ICCIS.2013.226
