Skip to main content
Springer Nature Link
Log in

A characterization of the existence of energies for neural networks

  • Learning, Coding, Robotics
  • Conference paper
  • First Online:
Automata, Languages and Programming (ICALP 1995)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 944))

Included in the following conference series:

  • 151 Accesses

Abstract

In this paper we give under an appropriate theoretical frame-work 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.

Support by the EC Working Group NeuroCOLT and French-Chile cooperation (ECOS-94)is aknowledged.

Partially supported by FONDECYT-94, EC-Chile project in applied mathematics and French-Chile cooperation (ECOS-94).

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Fogelman-Soulié, F., Goles, E., Weisbuch, G.: Transient length in sequential iteration of threshold functions. Disc. Appl. Math. 6, (1983) 95–98

    Google Scholar 

  2. Goles, E., Fogelman-Soulié, F., Pellegrin, D.: Decreasing energy functions as a tool for studying threshold netwoks. Disc. Appl. Math. 12, (1985) 261–277

    Google Scholar 

  3. Goles, E., Martinez, S.: Neural and automata networks. Math. and Applications, 58, Kluwer Acad. Pub., (1990)

    Google Scholar 

  4. Hopfield, J.J.: Neural networks and physical systems with emergent collective computational abilities. Proc. Natl. Acad. Sci., USA, 79, (1982) 2554–2558

    Google Scholar 

  5. Kobuchi, Y.: State evaluation functions and Lyapunov functions for neural networks. Neural Networks, 4, (1991) 505–510

    Google Scholar 

  6. X. Xu W.T. Tsai, Construction associative memories using neural networks. Neural Networks, 3, (1990) 301–309

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Laboratoire de l'Informatique du Parallélisme, CNRS Ecole Normale Supérieure de Lyon, 69364, Lyon, France

    Michel Cosnard & Eric Goles

  2. Departamento de Ingeniería Matemática, Universidad de Chile, casilla 170-3, Santiago, Chile

    Michel Cosnard & Eric Goles

Authors
  1. Michel Cosnard
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Eric Goles
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Zoltán Fülöp Ferenc Gécseg

Rights and permissions

Reprints and permissions

Copyright information

© 1995 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Cosnard, M., Goles, E. (1995). A characterization of the existence of energies for neural networks. In: Fülöp, Z., Gécseg, F. (eds) Automata, Languages and Programming. ICALP 1995. Lecture Notes in Computer Science, vol 944. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60084-1_106

Download citation

  • DOI: https://doi.org/10.1007/3-540-60084-1_106

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-60084-8

  • Online ISBN: 978-3-540-49425-6

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics