From Spin Glasses to Learning of Neural Networks

— The conceptual basics of spin glass theory are reviewed. A description of the mathematical apparatus developed for spin glasses and the model of the restricted Boltzmann machine (RBM) is presented. Optimization of the RBM learning algorithm using nongradient methods is explored. A method to extrac...

Full description

Saved in:
Bibliographic Details
Published in:Physics of particles and nuclei 2022-08, Vol.53 (4), p.834-847
Main Authors: Perepelkin, E. E., Sadovnikov, B. I., Inozemtseva, N. G., Rudamenko, R. A., Tarelkin, A. A., Sysoev, P. N., Polyakova, R. V., Sadovnikova, M. B.
Format: Article
Language:English
Subjects:
Particle and Nuclear Physics
Physics
Physics and Astronomy
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:— The conceptual basics of spin glass theory are reviewed. A description of the mathematical apparatus developed for spin glasses and the model of the restricted Boltzmann machine (RBM) is presented. Optimization of the RBM learning algorithm using nongradient methods is explored. A method to extract the learning algorithm hyperparameter, temperature, has been described and used. Critical phenomena in the RBM—entropy crisis, and difference between the temperatures of the learning sample creation and processing—are studied.
ISSN:1063-7796
1531-8559
DOI:10.1134/S1063779622040128