A spherical Hopfield model

DOI10.1088/0305-4470/36/41/002MaRDI QIDQ4468572zbMATH OpenOpenAlexFDO

Authors D. Bollé, Theodorus M. Nieuwenhuizen, Isaac Pérez Castillo, T. Verbeiren

Publication date 10 June 2004

Published in Journal of Physics A: Mathematical and General (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/cond-mat/0305316

Mathematics Subject Classification ID

Statistical thermodynamics (82B30) Neural nets applied to problems in time-dependent statistical mechanics (82C32) Neural networks for/in biological studies, artificial life and related topics (92B20)

Abstract: We introduce a spherical Hopfield-type neural network involving neurons and patterns that are continuous variables. We study both the thermodynamics and dynamics of this model. In order to have a retrieval phase a quartic term is added to the Hamiltonian. The thermodynamics of the model is exactly solvable and the results are replica symmetric. A Langevin dynamics leads to a closed set of equations for the order parameters and effective correlation and response function typical for neural networks. The stationary limit corresponds to the thermodynamic results. Numerical calculations illustrate our findings.

Recommendations

• Some rigorous results on the Hopfield neural network model
• Hopfield model with a dynamic threshold
• THE INVESTIGATION OF THE SPIN GLASS PROPERTIES OF THE HOPFIELD NEURAL NETWORK MODEL
• scientific article; zbMATH DE number 897979
• scientific article; zbMATH DE number 2201931

Cited in

(6)

• Stationary states of a spherical minority game with ergodicity breaking
• On the Marchenko–Pastur law in analog bipartite spin-glasses
• Legendre equivalences of spherical Boltzmann machines
• Aging in the retrieval phase of the Hopfield model
• Thermodynamics of bidirectional associative memories
• Statistical mechanics of vector Hopfield network near and above saturation

This page was built for publication: A spherical Hopfield model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4468572)