An artificial maximum neural network: a winner-take-all neuron model forcing the state of the system in a solution domain

Yoshiyasu Takefuji, Kuo Chun Lee, Hideo Also

Research output: Contribution to journalArticle

69 Citations (Scopus)

Abstract

A maximum neuron model is proposed in order to force the state of the system to converge to the solution in neural dynamics. The state of the system is always forced in a solution domain. The artificial maximum neural network is used for the module orientation problem and the bipartite subgraph problem. The usefulness of the maximum neural network is empirically demonstrated by simulating randomly generated massive nstances (examples) in both problems. In randomly generated more than one thousand instances our system always converges to the solution within one hundred iteration steps regardless of the problem size. Our simulation results show the effectiveness of our algorithms and support our claim that one class of NP-complete problems may be solvable in a polynomial time.

Original languageEnglish
Pages (from-to)243-251
Number of pages9
JournalBiological Cybernetics
Volume67
Issue number3
DOIs
Publication statusPublished - 1992 Jul
Externally publishedYes

Fingerprint

Neurons
Neural networks
Computational complexity
Polynomials

ASJC Scopus subject areas

  • Biophysics

Cite this

An artificial maximum neural network : a winner-take-all neuron model forcing the state of the system in a solution domain. / Takefuji, Yoshiyasu; Lee, Kuo Chun; Also, Hideo.

In: Biological Cybernetics, Vol. 67, No. 3, 07.1992, p. 243-251.

Research output: Contribution to journalArticle

@article{26d9c187411144fea4447ac64117bffb,
title = "An artificial maximum neural network: a winner-take-all neuron model forcing the state of the system in a solution domain",
abstract = "A maximum neuron model is proposed in order to force the state of the system to converge to the solution in neural dynamics. The state of the system is always forced in a solution domain. The artificial maximum neural network is used for the module orientation problem and the bipartite subgraph problem. The usefulness of the maximum neural network is empirically demonstrated by simulating randomly generated massive nstances (examples) in both problems. In randomly generated more than one thousand instances our system always converges to the solution within one hundred iteration steps regardless of the problem size. Our simulation results show the effectiveness of our algorithms and support our claim that one class of NP-complete problems may be solvable in a polynomial time.",
author = "Yoshiyasu Takefuji and Lee, {Kuo Chun} and Hideo Also",
year = "1992",
month = "7",
doi = "10.1007/BF00204397",
language = "English",
volume = "67",
pages = "243--251",
journal = "Biological Cybernetics",
issn = "0340-1200",
publisher = "Springer Verlag",
number = "3",

}

TY - JOUR

T1 - An artificial maximum neural network

T2 - a winner-take-all neuron model forcing the state of the system in a solution domain

AU - Takefuji, Yoshiyasu

AU - Lee, Kuo Chun

AU - Also, Hideo

PY - 1992/7

Y1 - 1992/7

N2 - A maximum neuron model is proposed in order to force the state of the system to converge to the solution in neural dynamics. The state of the system is always forced in a solution domain. The artificial maximum neural network is used for the module orientation problem and the bipartite subgraph problem. The usefulness of the maximum neural network is empirically demonstrated by simulating randomly generated massive nstances (examples) in both problems. In randomly generated more than one thousand instances our system always converges to the solution within one hundred iteration steps regardless of the problem size. Our simulation results show the effectiveness of our algorithms and support our claim that one class of NP-complete problems may be solvable in a polynomial time.

AB - A maximum neuron model is proposed in order to force the state of the system to converge to the solution in neural dynamics. The state of the system is always forced in a solution domain. The artificial maximum neural network is used for the module orientation problem and the bipartite subgraph problem. The usefulness of the maximum neural network is empirically demonstrated by simulating randomly generated massive nstances (examples) in both problems. In randomly generated more than one thousand instances our system always converges to the solution within one hundred iteration steps regardless of the problem size. Our simulation results show the effectiveness of our algorithms and support our claim that one class of NP-complete problems may be solvable in a polynomial time.

UR - http://www.scopus.com/inward/record.url?scp=0344194652&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344194652&partnerID=8YFLogxK

U2 - 10.1007/BF00204397

DO - 10.1007/BF00204397

M3 - Article

AN - SCOPUS:0344194652

VL - 67

SP - 243

EP - 251

JO - Biological Cybernetics

JF - Biological Cybernetics

SN - 0340-1200

IS - 3

ER -