Abstract
We study the set of fixed points of a Hopfield-type neural network with a connection matrix constructed from a high-symmetry set of memorized patterns using the Hebb rule. The memorized patterns depending on an external parameter are interpreted as distorted copies of a vector standard to be learned by the network. The dependence of the fixed-point set of the network on the distortion parameter is described analytically. The investigation results are interpreted in terms of neural networks and the Ising model.
Similar content being viewed by others
References
E. Domany, J. L. van Hemmen, and K. Schulten, eds.,Models of Neural Networks, Springer, Berlin (1991).
U. Blasum, W. Hochstattler, C. Moll, and H. Rieger, Preprint cond-mat/9604052 (1996); H. Rieger and U. Blasum, Preprint cond-mat/9608136 (1996).
N. Sourlas,Europhys. Lett.,25, 159 (1994).
G. Parisi, Preprint cond-mat/9412030 (1994).
E. M. Braverman and I. B. Muchnik,Structural Processing Methods for Empirical Data [in Russian], Nauka, Moscow (1983).
L. B. Litinskii,Mat. Modelirovanie,8, 11 (1996).
L. B. Litinsky, Preprint cond-mat/9706280 (1997).
J. J. Hopfield,Proc. Nat. Acad. Sci. USA,79, 2554 (1982).
P. Baldi,IEEE Trans. Inform. Theory,34, 523 (1988).
T. L. H. Watkin, A. Rau, and M. Biehl,Rev. Mod. Phys.,65, 499 (1993).
B. Muller, J. Reinhardt, and M. T. Strikland,Neural Networks: An Introduction (2nd ed.), Springer, Berlin (1995).
S. Diderich and M. Opper,Phys. Rev. Lett.,58, 949 (1987).
L. B. Litinskii,Theory. Math. Phys.,101, 1492 (1994).
A. A. Vedenov, A. A. Ezhov, et al., “Structure of patterns in associative memory models,” in:Neural Networks: Theory and Architecture (A. V. Holden and V. I. Kryukov, eds.), Manchester Univ. Press, Manchester (1991), p. 169.
L. B. Litinskii, “Taking the magnetic field for the Ising model with a high-symmetry connection matrix into account,” (in preparation).
L. B. Litinsky,Solid State Commun.,79, 879 (1991).
Author information
Authors and Affiliations
Additional information
Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 118, No. 1, pp. 133–158, January, 1999.
Rights and permissions
About this article
Cite this article
Litinskii, L.B. High-symmetry Hopfield-type neural networks. Theor Math Phys 118, 107–127 (1999). https://doi.org/10.1007/BF02557200
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02557200