Increasing attraction of pseudo-inverse autoassociative networks

被引：0

作者：

Gorodnichy, DO ^{[1
]}

Reznik, AM ^{[1
]}

机构：

[1] UNIV ALBERTA,DEPT COMP SCI,EDMONTON,AB T6G 2H1,CANADA

来源：

NEURAL PROCESSING LETTERS | 1997年 / 5卷 / 02期

关键词：

attraction radius; pseudo-inverse learning rule; self-connections;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We show howl partial reduction of self-connections of the network designed with the pseudo-inverse learning rule increases the direct attraction radius of the network. Theoretical formula is obtained. Data obtained by simulation are presented.

引用

页码：121 / 125

页数：5

共 6 条

[1] ATTRACTION RADII IN BINARY HOPFIELD NETS ARE HARD TO COMPUTE [J].

FLOREEN, P ;

ORPONEN, P .

NEURAL COMPUTATION, 1993, 5 (05) :812-821

[2]

GORODNICHY DO, 1996, HELNET 94 95 INT WOR, V1, P198

[3]

Hertz J., 1991, Introduction to the Theory of Neural Computation

[4] ASSOCIATIVE RECALL OF MEMORY WITHOUT ERRORS [J].

KANTER, I ;

SOMPOLINSKY, H .

PHYSICAL REVIEW A, 1987, 35 (01) :380-392

[5]

Kohonen T., 1984, SELF ORG ASS MEMORY

[6] COLLECTIVE COMPUTATIONAL PROPERTIES OF NEURAL NETWORKS - NEW LEARNING-MECHANISMS [J].

PERSONNAZ, L ;

GUYON, I ;

DREYFUS, G .

PHYSICAL REVIEW A, 1986, 34 (05) :4217-4228

← 1 →