English

Quantum field theories, Markov random fields and machine learning

Machine Learning 2022-03-30 v2 Disordered Systems and Neural Networks Statistical Mechanics High Energy Physics - Lattice

Abstract

The transition to Euclidean space and the discretization of quantum field theories on spatial or space-time lattices opens up the opportunity to investigate probabilistic machine learning within quantum field theory. Here, we will discuss how discretized Euclidean field theories, such as the ϕ4\phi^{4} lattice field theory on a square lattice, are mathematically equivalent to Markov fields, a notable class of probabilistic graphical models with applications in a variety of research areas, including machine learning. The results are established based on the Hammersley-Clifford theorem. We will then derive neural networks from quantum field theories and discuss applications pertinent to the minimization of the Kullback-Leibler divergence for the probability distribution of the ϕ4\phi^{4} machine learning algorithms and other probability distributions.

Cite

@article{arxiv.2110.10928,
  title  = {Quantum field theories, Markov random fields and machine learning},
  author = {Dimitrios Bachtis and Gert Aarts and Biagio Lucini},
  journal= {arXiv preprint arXiv:2110.10928},
  year   = {2022}
}
R2 v1 2026-06-24T07:03:48.951Z