English

Learning disentangled representation for classical models

Strongly Correlated Electrons 2022-07-01 v2 Statistical Mechanics

Abstract

Finding disentangled representation plays a predominant role in the success of modern deep learning applications, but the results lack a straightforward explanation. Here we apply the information bottleneck method and its β\beta-VAE implementation to find the disentangled low-dimensional representation of classical models. For the Ising model, our results reveal a deep connection between the disentangled features and the physical order parameters, and the widely-used Bernoulli decoder is found to be learning a mean-field Hamiltonian at fixed temperature. This analogy motivates us to extend the application of β\beta-VAE to more complex classical models with non-binary variables using different decoder neural network and propose a modified architecture β2\beta^2-VAE to enforce thermal fluctuations in generated samples. Our work provides a way to design novel physics-informed algorithm that can yield learned features in potential correspondence with real physical properties.

Keywords

Cite

@article{arxiv.2110.08082,
  title  = {Learning disentangled representation for classical models},
  author = {Dongchen Huang and Danqing Hu and Yi-feng Yang},
  journal= {arXiv preprint arXiv:2110.08082},
  year   = {2022}
}

Comments

10 pages, 7 figures, 2 tables

R2 v1 2026-06-24T06:55:13.428Z