Unified stabilization approach to principal and minor components extraction algorithms

被引:65
作者
Chen, TP [1 ]
Amari, S
机构
[1] Fudan Univ, Dept Math, Shanghai 200433, Peoples R China
[2] RIKEN, Brain Sci Inst, Wako, Saitama 3510198, Japan
关键词
principal component extraction; minor component extraction; singular value decomposition; stability of dynamical system; natural gradient;
D O I
10.1016/S0893-6080(01)00116-2
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Principal component and minor component extractions provide powerful techniques in many information processing fields. There have been proposed a number of algorithms for principal and minor component (or subspace) extraction, which have different dynamical behaviors. In this paper, we give rigorous stability analysis of these algorithms, obtaining a unified insight view on the dynamical behaviors of various algorithms. (C) 2001 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:1377 / 1387
页数:11
相关论文
共 20 条
[1]   Adaptive method of realizing natural gradient learning for multilayer perceptrons [J].
Amari, S ;
Park, H ;
Fukumizu, K .
NEURAL COMPUTATION, 2000, 12 (06) :1399-1409
[2]   Stability analysis of learning algorithms for blind source separation [J].
Amari, S ;
Chen, TP ;
Cichocki, A .
NEURAL NETWORKS, 1997, 10 (08) :1345-1351
[3]   Natural gradient works efficiently in learning [J].
Amari, S .
NEURAL COMPUTATION, 1998, 10 (02) :251-276
[4]  
Amari S., 1985, SPRINGER LECT NOTES, V28
[5]  
Amari S.I., 1997, Contemp. Math., V203, P81, DOI DOI 10.1090/C0NM/203/02554
[6]   NEURAL THEORY OF ASSOCIATION AND CONCEPT-FORMATION [J].
AMARI, SI .
BIOLOGICAL CYBERNETICS, 1977, 26 (03) :175-185
[7]   DYNAMIC-SYSTEMS THAT SORT LISTS, DIAGONALIZE MATRICES, AND SOLVE LINEAR-PROGRAMMING PROBLEMS [J].
BROCKETT, RW .
LINEAR ALGEBRA AND ITS APPLICATIONS, 1991, 146 :79-91
[8]   LEAST-SQUARES MATCHING PROBLEMS [J].
BROCKETT, RW .
LINEAR ALGEBRA AND ITS APPLICATIONS, 1989, 122 :761-777
[9]   A unified algorithm for principal and minor components extraction [J].
Chen, TP ;
Amari, SI ;
Lin, Q .
NEURAL NETWORKS, 1998, 11 (03) :385-390
[10]  
Chen TP, 1997, NEURAL PROCESS LETT, V5, P105