4.5 Review

Restricted Boltzmann machine: Recent advances and mean-field theory*

期刊

CHINESE PHYSICS B
卷 30, 期 4, 页码 -

出版社

IOP Publishing Ltd
DOI: 10.1088/1674-1056/abd160

关键词

restricted Boltzmann machine (RBM); machine learning; statistical physics

资金

  1. Comunidad de Madrid
  2. Complutense University of Madrid (Spain) through the Atracci 'on de Talento program [2019T1/TIC-13298]

向作者/读者索取更多资源

This review focuses on the application of restricted Boltzmann machines in statistical physics, discussing mean-field theory and phase diagram analysis of RBMs in machine learning. Recent works on mean-field based learning algorithms and reproducing aspects of the learning process from ensemble dynamics equations or linear stability arguments are also discussed.
This review deals with restricted Boltzmann machine (RBM) under the light of statistical physics. The RBM is a classical family of machine learning (ML) models which played a central role in the development of deep learning. Viewing it as a spin glass model and exhibiting various links with other models of statistical physics, we gather recent results dealing with mean-field theory in this context. First the functioning of the RBM can be analyzed via the phase diagrams obtained for various statistical ensembles of RBM, leading in particular to identify a compositional phase where a small number of features or modes are combined to form complex patterns. Then we discuss recent works either able to devise mean-field based learning algorithms; either able to reproduce generic aspects of the learning process from some ensemble dynamics equations or/and from linear stability arguments.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.5
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据