Summary
We study the thermodynamic properties of the Hopfield model of an autoassociative memory. IfN denotes the number of neurons andM (N) the number of stored patterns, we prove the following results: IfM/N↓ 0 asN↑ ∞, then there exists an infinite number of infinite volume Gibbs measures for all temperaturesT<1 concentrated on spin configurations that have overlap with exactly one specific pattern. Moreover, the measures induced on the overlap parameters are Dirac measures concentrated on a single point and the Gibbs measures on spin configurations are products of Bernoulli measures. IfM/N → α, asN↓∞ for α small enough, we show that for temperaturesT smaller than someT(α)<1, the induced measures can have support only on a disjoint union of balls around the previous points, but we cannot construct the infinite volume measures through convergent sequences of measures.
Article PDF
Similar content being viewed by others
References
[A] Amit, D.J.: Modelling brain: the world of attractor neural networks. Cambridge: Cambridge University Press 1989
[AGS1] Amit, D.J., Gutfreund, H., Sompolinsky, H.: Spin-glass models of neural networks. Phys. Rev. A32, 1007–1018 (1985)
[AGS2] Amit, D.J., Gutfreund, H., Sompolinsky, H.: Storing infinite numbers of patterns in a spin glass model of neural networks. Phys. Rev. Letts.55, 1530–1533 (1985)
[Bi] Billingsley, P.: Convergence of Probability Measures. New York: Wiley, 1968
[BG] Bovier A. and Gayrard, V.: Rigorous results on the thermodynamics of the dilute Hopfield model J. Stat. Phys.72, 79–112 (1993)
[Co] Comets, F.: Large deviation estimates for a conditional probability distribution. Applications to random Gibbs measures. Probab. Theory Relat. Fields80, 407–432 (1989)
[E] Ellis, R.S.: Entropy, large deviations, and statistical mechanics. Berlin Heidelberg New York: Springer 1985
[FK] Füredi Z. and Komlós, J.: The eigenvalues of random symmetric matrices. Combinatorica1, 233–241 (1981)
[FP1] Pastur L.A. and Figotin, A.L.: Exactly soluble model of a spin glass. Sov. J. Low Temp. Phys.3(6), 378–383 (1977)
[FP2] Pastur L.A. and Figotin, A.L.: On the theory of disordered spin systems. Theor. Math. Phys.35, 403–414 (1978)
[G] Gayrard, V.: The thermodynamic limit of theq-state Potts-Hopfield model with infinitely many patterns. J. Stat. Phys.68, 977–1011 (1992)
[Ge] Georgii, H.-O.: Gibbs measures and phase transitions. Berlin: Walter de Gruyter 1988
[H] van Hemmen, J.L.: Spin glass models of a neural network. Phys. Rev. A34, 3435–3445 (1986)
[Ho] Hopfield, J.J.: Neural networks and physical systems with emergent collective computational abilities. Proc. Natl. Acad. Sci. USA79, 2554–2558 (1982)
[HS] Stratonovich, R.L.: On a method of calculating quantum distribution functions, Doklady Akad. Nauk S.S.S.R.115, 1097 (1957) Soviet Phys. Doklady2, 416–419 (1958) (translation)
[K] Koch, H.: A free energy bound for the Hopfield model. J. Phys. A.: Math Gen.26, L353-L355 (1993)
[KPa] Komlos J. and Paturi, R.: Convergence results in a autoassociative memory model. Neural Networks1, 239–250 (1988)
[KP] Koch H. and Piasko, J.: Some rigorous results on the Hopfield neural network model. J. Stat. Phys.55, 903 (1989)
[L] Michel Ledoux and Michel Talagrand: Probability in Banach Spaces. Berlin Heidelberg New York: Springer 1991
[MPV] Mézard, M., Parisi G. and Virasoro, M.A.: Spin-glass theory and beyond. Singapore: World Scientific 1988
[N] Newman, Ch.M.: Memory capacity in neural network models: Rigorous lower bounds. Neural Networks1, 223–238 (1988)
[PST] Pastur, L. Shcherbina M. and Tirozzi, B.: The replica symmetric solution without the replica trick for the Hopfield model. J. Stat. Phys.74, 1161–1183 (1994)
[Ro] Robbins, H.: A remark on Stirling's formula. Am. Math. Montly62, 26–29 (1955)
[SK] Sherrington D. and Kirkpatrick, S.: Solvable model of a spin glass. Phys. Rev. Lett.35, 1792–1796 (1972)
[ST] Shcherbina M. and Tirozzi, B.: The free energy for a class of Hopfield models. J. Stat. Phys.72, 113–125 (1992)
[SeT] Scacciatelli E. and Tirozzi, B.: Fluctuation of the free energy in the Hopfield model. J. Stat. Phys.67, 981–1008 (1992)
Author information
Authors and Affiliations
Additional information
Work partially supported by the Commission of the European Communities under contract No. SC1-CT91-0695
Rights and permissions
About this article
Cite this article
Bovier, A., Gayrard, V. & Picco, P. Gibbs states of the Hopfield model in the regime of perfect memory. Probab. Th. Rel. Fields 100, 329–363 (1994). https://doi.org/10.1007/BF01193704
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01193704