Neural Processing Letters

, Volume 2, Issue 5, pp 28–31 | Cite as

Hebb-Hopfield neural networks based on one-dimensional sets of neuron states

  • Witali L. Dunin-Barkowski
  • Natali B. Osovets


Neural Networks (NN), which interconnection matrix is the Hebb matrix of Hopfield (HH) [2,3] are considered. Quasi-continuos sets of neuron states are being used for network matrix production. It is shown, that in this case minima of Hopfield energy are at the bottom of deep ditches, corresponding to the basic set of network activity states for the HH NN. The corresponding states can be made to be stable states of the network. When neuron threshold fatigue is introduced, depending of its recent activity state, the network activity becomes cyclic, moving with a constant rate in one of the two possible directions in the ring, depending on the initial conditions. The phenomena described present novel robust types of NN behavior, which have a high probability to be encountered in living neural systems.


Fatigue Neural Network Stable State Activity State Nonlinear Dynamics 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    D. Marr. Simple memory: A theory for archicortex,Phil. Trans. Roy. Soc., vol. B-262, pp. 23–81, 1971.ADSGoogle Scholar
  2. [2]
    J.J. Hopfield. Neural networks and physical systems with emergent collective computational abilities,Proc. Nal. Acad. Sci., vol. 79, pp. 2554–2558, 1982.ADSMathSciNetGoogle Scholar
  3. [3]
    J.J. Hopfield. Neurons with graded response have collective computational properties like those of two-state neurons,Proc. Nat. Acad. Sci., vol. 81, pp. 3088–3092, 1984.ADSGoogle Scholar
  4. [4]
    V.S. Dotsenko, L.B. Ioffe, M.V. Feigelman, M.V. Tsodyks. Statisticheskie modeli nejronnykh setej, In A. Vedenov, editor,Fizicheskie i matematicheskie modeli nejronnych setej, Moscow, VINITI, vol. 1, 1990 (in Russian).Google Scholar
  5. [5]
    W.L. Dunin-Barkowski. Konfiguracionnye generatory nejronnoj ritmiki,Biofizika, vol. 29, no. 5. pp. 899–902, 1984 (in Russian).Google Scholar
  6. [6]
    E.W. Dunina-Barkowska, W.L. Dunin-Barkowski. Analog/digital controversies in neural computations.Neural Network World, vol. 3, no. 4, p. 361–372, 1993.Google Scholar
  7. [7]
    W.L. Dunin-Barkowski, N.B Osovets. Neural network with a predetermined activity dynamics. - In:Optical Memory and Neural Networks ’94: Optical Neural Networks, A.L. Mikaelyan, ed., Proc. SPIE, vol. 2430, pp. 118–127, 1994.Google Scholar
  8. [8]
    Y.I. Arshavskij, M.B. Berkinblit, W.L. Dunin-Barkowski. Rasprostranenie impulsov v koltse vozbudimoj tkani.Biofizika, vol. 10, no. 6, pp. 1048–1053, 1965 (in Russian).Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Witali L. Dunin-Barkowski
    • 1
  • Natali B. Osovets
    • 1
  1. 1.A.B. Kogan Research Institute for NeurocyberneticsRostov State UniversityRostov-on-DonRussia

Personalised recommendations