Stochastic normalizing flows as non-equilibrium transformations
This work addresses sampling efficiency for researchers in lattice field theories, but it appears incremental as it connects existing frameworks without introducing a new method.
The authors tackled the problem of inefficient sampling in lattice field theories by showing that stochastic normalizing flows align with out-of-equilibrium simulations based on Jarzynski's equality, and they developed an optimization strategy for these generative models with application examples.
Normalizing flows are a class of deep generative models that provide a promising route to sample lattice field theories more efficiently than conventional Monte Carlo simulations. In this work we show that the theoretical framework of stochastic normalizing flows, in which neural-network layers are combined with Monte Carlo updates, is the same that underlies out-of-equilibrium simulations based on Jarzynski's equality, which have been recently deployed to compute free-energy differences in lattice gauge theories. We lay out a strategy to optimize the efficiency of this extended class of generative models and present examples of applications.