THE UD RLS ALGORITHM FOR TRAINING FEEDFORWARD NEURAL NETWORKS

被引：0
作者：
Bilski, Jaroslaw [1 ]
机构：
[1] Czestochowa Tech Univ, Dept Comp Engn, Ul Armii Krajowej 36, PL-42200 Czestochowa, Poland
来源：
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS AND COMPUTER SCIENCE | 2005年 / 15卷 / 01期
关键词：
neural networks; learning algorithms; recursive least squares method; UD factorization;
D O I：
暂无
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
A new algorithm for training feedforward multilayer neural networks is proposed. It is based on recursive least squares procedures and U-D factorization, which is a well-known technique in filter theory. It will be shown that due to the U-D factorization method, our algorithm requires fewer computations than the classical RLS applied to feedforward multilayer neural network training.
引用
收藏
页码：115 / 123
页数：9
相关论文
共 25 条
[1]   A fast feedforward training algorithm using a modified form of the standard backpropagation algorithm [J].
Abid, S ;
Fnaiech, F ;
Najim, M .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (02) :424-430
[2]   Two highly efficient second-order algorithms for training feedforward networks [J].
Ampazis, N ;
Perantonis, SJ .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (05) :1064-1074
[3]  
[Anonymous], ARTIFICIAL NEURAL NE
[4]   A fast training algorithm for neural networks [J].
Bilski, J ;
Rutkowski, L .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING, 1998, 45 (06) :749-753
[5]  
Bilski J., 1995, THESIS AGH U SCI TEC
[6]  
Bilski J., 1996, 2 C NEUR NETW THEIR, P25
[7]  
Bishop C. M., 1996, Neural networks for pattern recognition
[8]  
Chen X.-H. Y. G.-A., 1992, NEURAL NETWORKS, V10, P517
[9]   Speeding up backpropagation algorithms by using Cross-Entropy combined with Pattern Normalization [J].
Joost, M ;
Schiffmann, W .
INTERNATIONAL JOURNAL OF UNCERTAINTY FUZZINESS AND KNOWLEDGE-BASED SYSTEMS, 1998, 6 (02) :117-126
[10]   EFFICIENT LEARNING ALGORITHMS FOR NEURAL NETWORKS (ELEANNE) [J].
KARAYIANNIS, NB ;
VENETSANOPOULOS, AN .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1993, 23 (05) :1372-1383
123