High capacity recurrent associative memories

被引：21

作者：

Davey, N ^{[1
]}

Hunt, SP ^{[1
]}

Adams, RG ^{[1
]}

机构：

[1] Univ Hertfordshire, Dept Comp Sci, Hatfield AL10 9AB, Herts, England

来源：

NEUROCOMPUTING | 2004年 / 62卷

关键词：

associative memory; attractor basins; Hopfield neural networks; learning rules; perceptron; performance measures; pseudo-inverse;

D O I：

10.1016/j.neucom.2004.02.007

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Various algorithms for constructing weight matrices for Hopfield-type associative memories are reviewed, including ones with much higher capacity than the basic model. These alternative. algorithms either iteratively approximate the projection weight matrix or use simple perceptron learning. An experimental investigation of the performance of networks trained by these algorithms is presented, including measurements of capacity, training time and their ability to correct corrupted versions of the training patterns. (C) 2004 Elsevier B.V. All rights reserved.

引用

页码：459 / 491

页数：33

共 38 条

[1] Learning in neural network memories [J].

Abbott, L. F. .

NETWORK-COMPUTATION IN NEURAL SYSTEMS, 1990, 1 (01) :105-122

[2] UNIVERSALITY IN THE SPACE OF INTERACTIONS FOR NETWORK MODELS [J].

ABBOTT, LF ;

KEPLER, TB .

JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL, 1989, 22 (12) :2031-2038

[3]

Amit D., 1989, Modelling Brain Function: the World of Attractor Neural Networks

[4] A comparative study of two learning rules for associative memory [J].

Athithan, G .

PRAMANA-JOURNAL OF PHYSICS, 1995, 45 (06) :569-582

[5] NEURAL NETWORKS - A LOCAL LEARNING PRESCRIPTION FOR ARBITRARY CORRELATED PATTERNS [J].

BLATT, MG ;

VERGINI, EG .

PHYSICAL REVIEW LETTERS, 1991, 66 (13) :1793-1796

[6]

BRUCOLI M, 1995, 38 MIDW S CIRC SYST

[7] Modeling memory: What do we learn from attractor neural networks? [J].

Brunel, N ;

Nadal, JP .

COMPTES RENDUS DE L ACADEMIE DES SCIENCES SERIE III-SCIENCES DE LA VIE-LIFE SCIENCES, 1998, 321 (2-3) :249-252

[8] Using features for the storage of patterns in a fully connected net [J].

Coombes, S ;

Taylor, JG .

NEURAL NETWORKS, 1996, 9 (05) :837-844

[9] GEOMETRICAL AND STATISTICAL PROPERTIES OF SYSTEMS OF LINEAR INEQUALITIES WITH APPLICATIONS IN PATTERN RECOGNITION [J].

COVER, TM .

IEEE TRANSACTIONS ON ELECTRONIC COMPUTERS, 1965, EC14 (03) :326-&

[10]

DAVEY N, 2000, 2 INT ICSC S NEUR CO

← 1 2 3 4 →