Convergence of On-Line Gradient Methods for Two-Layer Feedforward Neural Networks
Received:December 27, 1999  
Key Words: on-line gradient method   feedforward neural network   convergence.  
Fund Project:Supported by the Natural Science Foundation of China
Author NameAffiliation
LI Zheng-xue Dept. of Math., Jilin University, Changchun 130023, China 
WU Wei Dept. of Math., Dalian University of Technology, Dalian 116024, China 
ZHANG Hong-wei Dept. of Math., Dalian University of Technology, Dalian 116024, China 
Hits: 3526
Download times: 1987
Abstract:
      A discussion is given on the convergence of the on-line gradient methods for two-layer feedforward neural networks in general cases. The theories are applied to some usual activation functions and energy functions.
Citation:
DOI:10.3770/j.issn:1000-341X.2001.02.012
View Full Text  View/Add Comment  
  Copyright © Journal of Mathematical Research with Applications
Sponsored by：Dalian University of Technology, China Society for Industrial and Applied Mathematics
Address：No.2 Linggong Road, Ganjingzi District, Dalian City, Liaoning Province, P. R. China , Postal code ：116024
Service Tel：86-411-84707392 Email：jmre@dlut.edu.cn
Designed by Beijing E-tiller Co.,Ltd.
Due to security factors, early browsers cannot log in. It is recommended to log in with IE10, IE11, Google, Firefox, and 360 new browsers.