Serial updating rule for blind separation derived from the method of scoring

被引：27
作者：
Yang, HH [1 ]
机构：
[1] Oregon Grad Inst Sci & Technol, Dept Elect & Comp Engn, Beaverton, OR 97006 USA
[2] Oregon Grad Inst Sci & Technol, Dept Comp Engn & Sci, Beaverton, OR 97006 USA
来源：
IEEE TRANSACTIONS ON SIGNAL PROCESSING | 1999年 / 47卷 / 08期
关键词：
adaptive signal detection; gradient methods; maximum likelihood estimation; signal analysis; statistics;
D O I：
10.1109/78.774771
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
In the context of blind source separation, the method of scoring based on the inverse of the Fisher information matrix (FIM) becomes the serial updating; learning rule with equivariant property. This learning rule can be simplified to a low-complexity algorithm by using the asymptotic form of the FIM around the equilibrium. The simplified learning rule is still general enough to include some existing equivariant blind separation algorithms as its special cases.
引用
收藏
页码：2279 / 2285
页数：7
相关论文
共 9 条
[1]  
Amari S, 1996, ADV NEUR IN, V8, P757
[2]   Natural gradient works efficiently in learning [J].
Amari, S .
NEURAL COMPUTATION, 1998, 10 (02) :251-276
[3]  
Amari S, 1997, ADV NEUR IN, V9, P127
[4]   Infomax and maximum likelihood for blind source separation [J].
Cardoso, JF .
IEEE SIGNAL PROCESSING LETTERS, 1997, 4 (04) :112-114
[5]   Equivariant adaptive source separation [J].
Cardoso, JF ;
Laheld, BH .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1996, 44 (12) :3017-3030
[6]  
Horn R.A., 1991, TOPICS MATRIX ANAL
[7]  
Kay S. M., 1998, Fundamentals of Statistical Signal Processing, Volume 1:Estimation Theory, V1
[8]  
Pham DT, 1996, IEEE T SIGNAL PROCES, V44, P2768, DOI 10.1109/78.542183
[9]   Adaptive online learning algorithms for blind separation: Maximum entropy and minimum mutual information [J].
Yang, HH ;
Amari, S .
NEURAL COMPUTATION, 1997, 9 (07) :1457-1482
1