Mean-field theory for a spin-glass model of neural networks: TAP free energy and the paramagnetic to spin-glass transition

DOI10.1088/0305-4470/30/23/009zbMATH Open0932.82031arXivcond-mat/9705015OpenAlexW1970944708MaRDI QIDQ4255646FDOQ4255646

Authors: Kazuo Nakanishi, Hajime Takayama 

Publication date: 19 August 1999

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

Abstract: An approach is proposed to the Hopfield model where the mean-field treatment is made for a given set of stored patterns (sample) and then the statistical average over samples is taken. This corresponds to the approach made by Thouless, Anderson and Palmer (TAP) to the infinite-range model of spin glasses. Taking into account the fact that in the Hopfield model there exist correlations between different elements of the interaction matrix, we obtain its TAP free energy explicitly, which consists of a series of terms exhibiting the cluster effect. Nature of the spin-glass transition in the model is also examined and compared with those given by the replica method as well as the cavity method.

Full work available at URL: https://arxiv.org/abs/cond-mat/9705015

Recommendations

• THE INVESTIGATION OF THE SPIN GLASS PROPERTIES OF THE HOPFIELD NEURAL NETWORK MODEL
• The spin-glass phase-transition in the Hopfield model with \(p\)-spin interactions
• On the free energy of the Hopfield model

zbMATH Keywords

Hopfield modelreplica methodcavity methodspin-glass transitioninfinite-range model of spin glasses

Mathematics Subject Classification ID

Statistical mechanics of random media, disordered materials (including liquid crystals and spin glasses) (82D30) Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Cited In (8)

• A mean field model for spin glasses based on a 2-level perceptron
• THE INVESTIGATION OF THE SPIN GLASS PROPERTIES OF THE HOPFIELD NEURAL NETWORK MODEL
• The simulation of spin glass model of neural networks by the Wang-Landau algorithm
• Black holes and the loss landscape in machine learning
• Expansion of the effective action around non-Gaussian theories
• High-temperature expansions and message passing algorithms
• TAP free energy, spin glasses and variational inference
• Probabilistic data modelling with adaptive TAP mean-field theory

This page was built for publication: Mean-field theory for a spin-glass model of neural networks: TAP free energy and the paramagnetic to spin-glass transition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4255646)