English

Boltzmann machines as two-dimensional tensor networks

Statistical Mechanics 2021-09-01 v1 Machine Learning Computational Physics Quantum Physics Machine Learning

Abstract

Restricted Boltzmann machines (RBM) and deep Boltzmann machines (DBM) are important models in machine learning, and recently found numerous applications in quantum many-body physics. We show that there are fundamental connections between them and tensor networks. In particular, we demonstrate that any RBM and DBM can be exactly represented as a two-dimensional tensor network. This representation gives an understanding of the expressive power of RBM and DBM using entanglement structures of the tensor networks, also provides an efficient tensor network contraction algorithm for the computing partition function of RBM and DBM. Using numerical experiments, we demonstrate that the proposed algorithm is much more accurate than the state-of-the-art machine learning methods in estimating the partition function of restricted Boltzmann machines and deep Boltzmann machines, and have potential applications in training deep Boltzmann machines for general machine learning tasks.

Keywords

Cite

@article{arxiv.2105.04130,
  title  = {Boltzmann machines as two-dimensional tensor networks},
  author = {Sujie Li and Feng Pan and Pengfei Zhou and Pan Zhang},
  journal= {arXiv preprint arXiv:2105.04130},
  year   = {2021}
}

Comments

12 pages, 11 figures

R2 v1 2026-06-24T01:55:51.407Z