"Machine Learning for Thermodynamic Observables in Lattice Field Theories"
Abstract: In this talk, I will discuss how applying machine learning
techniques to lattice field theory is a promising route for solving
problems where Markov Chain Monte Carlo (MCMC) methods are problematic.
More specifically, I will show that deep generative models can be used to
estimate thermodynamic observables like the free energy, which contrasts
with existing MCMC-based methods that are limited to only estimate free
energy differences. I will demonstrate the effectiveness of the proposed
method for two-dimensional $\phi^4$ theory and compare it to MCMC-based
methods in detailed numerical experiments. This talk is based on work
with Kim Nicoli and others, PRL 126 (2021) 032001.