Static and dynamic attractors of autoassociative neural networks

  • Dmitry O. Gorodnichy
  • Alexandre M. Reznik
Poster Session C: Compression, Hardware & Software, Image Databases, Neural Network, Object Recognition & Reconstruction
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1311)


In this paper we study the problem of the occurrence of cycles in autoassociative neural networks. We call these cycles dynamic attractors, show when and why they occur and how they can be identified. Of particular interest is the pseudo-inverse network with reduced self-connection. We prove that it has dynamic attractors, which occur with a probability proportional to the number of prototypes and the degree of weight reduction. We show how to predict and avoid them.


pattern recognition neural network pseudo-inverse rule stable state 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    S. Amari. Neural theory of association and concept formation, Biological Cybernetics, vol 26, pp. 175–185, 1977.CrossRefPubMedGoogle Scholar
  2. 2.
    T. Kohonen. Self-organization and associative memory, Berlin:Springler, 1984.Google Scholar
  3. 3.
    J.J. Hopfield. Neural networks and physical systems with emergent collective computational abilities, Proc. Nat. Acad. Sci. USA, vol 79, pp. 2554–2558, 1982.PubMedGoogle Scholar
  4. 4.
    L. Personnaz, I. Guyon and G. Dreyfus. Collective computational properties of neural networks: New learning mechanisms, Phys. Rev. A vol 34, pp. 4217–4228, 1986.CrossRefPubMedGoogle Scholar
  5. 5.
    M.A. Arbib and A.R. Hanson. Vision, Brain, and Cooperative Computation, The MIT Press, 1987.Google Scholar
  6. 6.
    Y. Zhou and R. Chellappa. Artificial Neural Networks for Computer Vision, Research Notes in Neural Computing, vol 5, Springer-Verlag, New-York, 1992.Google Scholar
  7. 7.
    D. Valentin, H. Abdi, A.J. O'Toole, G.W. Cottrell. Connectionist models of face processing: A survey. Pattern Recognition, vol 27, 1208–1230, 1994.CrossRefGoogle Scholar
  8. 8.
    V.A. Morozov. Methods for Solving Incorrectly Posed Problems, Springer-Verlag, New-York, 1984.Google Scholar
  9. 9.
    D.O. Gorodnichy, W.W. Armstrong, X. Li. Adaptive Logic Networks for Facial Feature Detection, in these proceedings of ICIAP'97.Google Scholar
  10. 10.
    K. Cheung, L. Atlas, R. Marks. Synchronous vs asynchronous behavior of Hopfield's CAM neural net, Applied Optics, Vol. 26, No 22, pp. 4808–4813, 1987.Google Scholar
  11. 11.
    E. Goles, F. Fogelman, D. Pellegrin. Decreasing Energy Functions as a Tool for Studying Threshold Networks. Discrete Appl. Math., no 12, pp. 261–277, 1985.Google Scholar
  12. 12.
    D.O. Gorodnichy. Desaturating Coefficient for Projection Learning Rule, Intern. Conf. on Artificial Neural Networks (ICANN'96) Proceedings, pp. 469–476, Bochum, Germany, Springer-Verlag, Lecture Notes in Computer Science 1112, 1996.Google Scholar
  13. 13.
    D.O. Gorodnichy and A.M. Reznik. Increasing Attraction of Pseudo-Inverse Autoassociative Networks, Neural Processing Letters, volume 5, issue 2, pp. 123–127, 1997.CrossRefGoogle Scholar
  14. 14.
    A.M. Reznik. Iterativnyî proekcionnyî algoritm obucheniya neîronnykh seteî Kibernetika i sistemnyî analiz, in russian, (Peзниκ A. M. Иτepaтивый пpoeкцeмный aлгopитм oбyчeния нeйpoнных ceтeй, ϰибepнeтикa и cиcтeмный aнaлиз) no 6, pp. 131–141, 1993.Google Scholar
  15. 15.
    A. Frumkin, E. Moses. Physicality of the Little model, Phys. Rev. A vol 34, No 1, pp. 714–716, July 1986.CrossRefPubMedGoogle Scholar
  16. 16.
    A. Schultz. Five variations of Hopfield Associative memory networks, Journal of Artificial Neural Networks vol. 2, no 3, pp. 285–294, 1995.Google Scholar
  17. 17.
    D.O. Gorodnichy, A.M. Reznik. NEUTRAM — A Transputer Based Neural Network Simulator, Proc. of Second Intern. Conf. on Software for Multiprocessors and Supercomputers Theory, Practice, Experience (SMS TPE'94), pp. 136-142, Moscow, Russia, 1994.Google Scholar
  18. 18.
    I. Kanter and H. Sampolinsky. Associative recall of memory without errors, Phys. Rev. A vol 35, pp. 380–392, 1987.CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Dmitry O. Gorodnichy
    • 1
  • Alexandre M. Reznik
    • 2
  1. 1.Dept. of Computing ScienceUniversity of AlbertaEdmontonCanada
  2. 2.IPMMSCybernetics Center of Ukrainian Ac.Sc.KievUkraine

Personalised recommendations