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
[11]   NEUROGENETIC LEARNING - AN INTEGRATED METHOD OF DESIGNING AND TRAINING NEURAL NETWORKS USING GENETIC ALGORITHMS [J].
KITANO, H .
PHYSICA D-NONLINEAR PHENOMENA, 1994, 75 (1-3) :225-238
[12]   Neighborhood based Levenberg-Marquardt algorithm for neural network training [J].
Lera, G ;
Pinzolas, M .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (05) :1200-1203
[13]  
Leung CS, 2001, IEEE T NEURAL NETWOR, V12, P1314, DOI 10.1109/72.963768
[14]  
Liou R. J., 1992, IEEE T SIGNAL PROCES, V40, P443
[15]   A SCALED CONJUGATE-GRADIENT ALGORITHM FOR FAST SUPERVISED LEARNING [J].
MOLLER, MF .
NEURAL NETWORKS, 1993, 6 (04) :525-533
[16]   Fast second order learning algorithm for feedforward multilayer neural networks and its applications [J].
Osowski, S ;
Bojarczak, P ;
Stodolski, M .
NEURAL NETWORKS, 1996, 9 (09) :1583-1596
[17]   AN EFFICIENT CONSTRAINED LEARNING ALGORITHM WITH MOMENTUM ACCELERATION [J].
PERANTONIS, SJ ;
KARRAS, DA .
NEURAL NETWORKS, 1995, 8 (02) :237-249
[18]  
Rutkowski L., 1994, ADAPTIVE SIGNAL PROC
[19]  
Rutkowski L., 2003, FAMILY RLS NEURAL NE
[20]  
STROBACH P, 1990, LINEAR PREDICTION TH
123