Speaker
Description
In recent years, the combination of neural network architectures and out-of-equilibrium methods has emerged as a powerful framework to build a new generation of numerical algorithms in various scientific fields. In lattice field theory, a prominent example is Stochastic Normalizing Flows, which have been used to tackle topological freezing in $\textrm{SU}(3) $ gauge theory. In this talk, we briefly introduce out-of-equilibrium sampling methods and how to enhance them with neural networks. We then discuss a general approach for the large-scale training of Neural-Enhanced Out-of-equilibrium (NEO) methods based on Parallel Tempering and well-suited loss functions, showcasing numerical results for the $\textrm{O}(3)$ non-linear sigma model. We conclude by discussing potential future directions for NEO methods, highlighting the search for novel scenarios or algorithmic synergies, such as multiscale approaches, where these algorithms can enable novel simulation strategies.