Machine learning with quantum field theories

29 Jul 2021, 05:00
15m
Oral presentation Algorithms (including Machine Learning, Quantum Computing, Tensor Networks) Algorithms (including Machine Learning, Quantum Computing, Tensor Networks)

Speaker

Dimitrios Bachtis

Description

The precise equivalence between discretized Euclidean field theories and a certain class of probabilistic graphical models, namely the mathematical framework of Markov random fields, opens up the opportunity to investigate machine learning from the perspective of quantum field theory. In this talk we will demonstrate, through the Hammersley-Clifford theorem, that the ϕ4 scalar field theory on a square lattice satisfies the local Markov property and can therefore be recast as a Markov random field. We will then derive from the ϕ4 theory machine learning algorithms and neural networks which can be viewed as generalizations of conventional neural network architectures. Finally, we will conclude by presenting applications based on the minimization of an asymmetric distance between the probability distribution of the ϕ4 machine learning algorithms and that of target probability distributions.

Authors

Dimitrios Bachtis Gert Aarts (Swansea University) Biagio Lucini (Swansea University)

Presentation materials