Title: Assisting Sampling with Learning: Adaptive Monte Carlo with Normalizing Flows
Abstract: In many applications in computational sciences and statistical inference, one seeks to compute expectations on complex high-dimensional distributions. These problems are often plagued by multi-modality/metastability; slow relaxation between unconnected modes leads to slow convergence of estimators of such expectations. In this talk, I will present a strategy to enhance sampling with deep generative models called Normalizing Flows. We will see when simultaneously sampling and training can incur a drastic acceleration of MCMC convergence and discuss current limits in learning for sampling.
This is joint work with Grant Rotskoff (Stanford) and Eric Vanden-Eijnden (Courant Institute, NYU).
If you would like to attend, please email firstname.lastname@example.org for the Zoom link.