The number of equilibrium points and their basins of attraction of neural networks with a kind of cyclic connection matrix

Norikazu Takahashi; Tetsuo Nishi

The number of equilibrium points and their basins of attraction of neural networks with a kind of cyclic connection matrix

Research output: Contribution to journal › Article › peer-review

Abstract

A method of synthesis of mutually coupled neural networks possessing many equilibrium points whose basins of the attraction are of almost the same size is given. The connection matrix of the network is a cyclic matrix with all diagonal elements vanishing.

Original language	English
Pages (from-to)	101-104
Number of pages	4
Journal	Research Reports on Information Science and Electrical Engineering of Kyushu University
Volume	1
Issue number	1
Publication status	Published - Sept 1 1996
Externally published	Yes

Keywords

Basin of attraction
Cyclic matrix
Hopfield neural networks
Memory capacity

ASJC Scopus subject areas

Computer Science(all)
Electrical and Electronic Engineering

Cite this

@article{1373ed8a290d42369b39c14a966ba667,

title = "The number of equilibrium points and their basins of attraction of neural networks with a kind of cyclic connection matrix",

abstract = "A method of synthesis of mutually coupled neural networks possessing many equilibrium points whose basins of the attraction are of almost the same size is given. The connection matrix of the network is a cyclic matrix with all diagonal elements vanishing.",

keywords = "Basin of attraction, Cyclic matrix, Hopfield neural networks, Memory capacity",

author = "Norikazu Takahashi and Tetsuo Nishi",

year = "1996",

month = sep,

day = "1",

language = "English",

volume = "1",

pages = "101--104",

journal = "Research Reports on Information Science and Electrical Engineering of Kyushu University",

issn = "1342-3819",

publisher = "Kyushu University, Faculty of Science",

number = "1",

}

TY - JOUR

T1 - The number of equilibrium points and their basins of attraction of neural networks with a kind of cyclic connection matrix

AU - Takahashi, Norikazu

AU - Nishi, Tetsuo

PY - 1996/9/1

Y1 - 1996/9/1

N2 - A method of synthesis of mutually coupled neural networks possessing many equilibrium points whose basins of the attraction are of almost the same size is given. The connection matrix of the network is a cyclic matrix with all diagonal elements vanishing.

AB - A method of synthesis of mutually coupled neural networks possessing many equilibrium points whose basins of the attraction are of almost the same size is given. The connection matrix of the network is a cyclic matrix with all diagonal elements vanishing.

KW - Basin of attraction

KW - Cyclic matrix

KW - Hopfield neural networks

KW - Memory capacity

UR - http://www.scopus.com/inward/record.url?scp=0030234221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030234221&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0030234221

SN - 1342-3819

VL - 1

SP - 101

EP - 104

JO - Research Reports on Information Science and Electrical Engineering of Kyushu University

JF - Research Reports on Information Science and Electrical Engineering of Kyushu University

IS - 1

ER -

The number of equilibrium points and their basins of attraction of neural networks with a kind of cyclic connection matrix

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this