A theory of solving TAP equations for Ising models with general invariant random matrices

DOI10.1088/1751-8113/49/11/114002zbMath1342.82042arXiv1509.01229OpenAlexW2228653206WikidataQ41874856 ScholiaQ41874856MaRDI QIDQ2994707

Burak Çakmak, Ole Winther, Manfred Opper

Publication date: 3 August 2016

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

Full work available at URL: https://arxiv.org/abs/1509.01229

zbMATH Keywords

random matricesIsing modelsfree probabilityTAP equationsdynamical functional theoryiterative convergent algorithms

Mathematics Subject Classification ID

Random matrices (probabilistic aspects) (60B20) Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics (82B20)

Related Items (10)

Approximate message passing algorithms for rotationally invariant matrices ⋮ Mean-field inference methods for neural networks ⋮ Analysis of Bayesian inference algorithms by the dynamical functional approach ⋮ Analysis of random sequential message passing algorithms for approximate inference ⋮ Perturbative construction of mean-field equations in extensive-rank matrix factorization and denoising ⋮ A statistical physics approach to learning curves for the inverse Ising problem ⋮ High-temperature expansions and message passing algorithms ⋮ Mean-field theory of graph neural networks in graph partitioning ⋮ A dynamical mean-field theory for learning in restricted Boltzmann machines ⋮ A Unifying Tutorial on Approximate Message Passing

This page was built for publication: A theory of solving TAP equations for Ising models with general invariant random matrices