A characterization of the existence of energies for neural networks

Michel Cosnard, Eric Goles

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper we give under an appropriate theoretical framework a characterization about neural networks which admit an energy. We prove that a neural network admits an energy if and only if the weight matrix verifies two conditions: The diagonal elements are non-negative and the associated incidence graph does not admit non-quasi-symmetric circuits.

Original languageEnglish
Title of host publicationAutomata, Languages and Programming - 22nd International Colloquium, ICALP 1995, Proceedings
EditorsZoltan Fulop, Ferenc Gecseg
PublisherSpringer Verlag
Pages570-580
Number of pages11
ISBN (Print)3540600841, 9783540600848
DOIs
StatePublished - 1995
Externally publishedYes
Event22nd International Colloquium on Automata, Languages and Programming, ICALP 1995 - Szeged, Hungary
Duration: 10 Jul 199514 Jul 1995

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume944
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference22nd International Colloquium on Automata, Languages and Programming, ICALP 1995
Country/TerritoryHungary
CitySzeged
Period10/07/9514/07/95

Fingerprint

Dive into the research topics of 'A characterization of the existence of energies for neural networks'. Together they form a unique fingerprint.

Cite this