UVictoria Probability and Dynamics Seminar: Garrett Mulcahy

This talk will present a novel discretization scheme for Wasserstein gradient flows (i.e. curves of steepest descent for functions of probability measures) based on the successive computation of Schrödinger bridges with equal marginals. In doing so, we develop a small-time approximation of same-marginal Schrödinger bridges using Langevin diffusion. This talk will begin with an introduction to entropic regularized optimal transport, introducing all the notions essential for understanding this scheme. Based on joint work with Medha Agarwal, Zaid Harchaoui, and Soumik Pal.