QUALITY AND EFFICIENCY OF RETRIEVAL FOR WILLSHAW-LIKE AUTOASSOCIATIVE NETWORKS .1. CORRECTION

被引：3

作者：

FROLOV, A

KARTASHOV, A

GOLTSEV, A

FOLK, R

机构：

来源：

NETWORK-COMPUTATION IN NEURAL SYSTEMS | 1995年 / 6卷 / 04期

关键词：

D O I：

10.1088/0954-898X/6/4/001

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The informational properties of a neural network model of an autoassociative memory based on binary Hebbian synapses are investigated. The model is a modification of the Willshaw network with a floating threshold which keeps approximately constant the number of active neurons (winners) at each time step. In the asymptotic case of large number of neurons, informational characteristics have been calculated analytically for single-step correction. Comparison with simulations shows that the maximal correction efficiency attains its asymptotic values for networks with surprisingly small number of neurons. Simulation results for multistep correction show considerable improvement over the single-step case.

引用

页码：513 / 534

页数：22

共 32 条

[11]

Frolov A.A., 1987, NEURAL MODELS ASSOCI

[12] INFORMATIONAL CHARACTERISTICS OF NEURAL NETWORKS CAPABLE OF ASSOCIATIVE LEARNING BASED ON HEBBIAN PLASTICITY [J].

FROLOV, AA ;

MURAVEV, IP .

NETWORK-COMPUTATION IN NEURAL SYSTEMS, 1993, 4 (04) :495-536

[13]

FROLOV AA, 1989, BIOFIZIKA+, V34, P868

[14]

FROLOV AA, 1996, UNPUB NETWORK COMPUT

[15]

GARDNER E, 1986, J PHYS A-MATH GEN, V19, pL107

[16] RECALL OF EVENTS THROUGH LEARNING OF ASSOCIATIONS BETWEEN THEIR PARTS [J].

GARDNERMEDWIN, AR .

PROCEEDINGS OF THE ROYAL SOCIETY SERIES B-BIOLOGICAL SCIENCES, 1976, 194 (1116) :375-402

[17] STATISTICAL-ANALYSIS OF THE DYNAMICS OF A SPARSE ASSOCIATIVE MEMORY [J].

GIBSON, WG ;

ROBINSON, J .

NEURAL NETWORKS, 1992, 5 (04) :645-661

[18] PARALLEL DYNAMICS OF THE NEURAL NETWORK WITH THE PSEUDOINVERSE COUPLING MATRIX [J].

HENKEL, RD ;

OPPER, M .

JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL, 1991, 24 (09) :2201-2218

[19] NEURAL NETWORKS AND PHYSICAL SYSTEMS WITH EMERGENT COLLECTIVE COMPUTATIONAL ABILITIES [J].

HOPFIELD, JJ .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1982, 79 (08) :2554-2558

[20] NEURAL NETWORKS WITH LOW-LEVELS OF ACTIVITY - ISING VS MCCULLOCH-PITTS NEURONS [J].

HORNER, H .

ZEITSCHRIFT FUR PHYSIK B-CONDENSED MATTER, 1989, 75 (01) :133-136

← 1 2 3 4 →