Neural Optimal Transport Methods for Unpaired Learning

Alexander will talk about the recent techniques for learning deep generative models based on the mathematical theory of optimal transport (OT). The main focus will be on the novel neural-network-based methods for computing continuous OT to solve domain translation problems (style transfer, super-resolution, modality transfer, etc.) given only unpaired datasets.
Join online: https://www.youtube.com/watch?v=CwV1oxIE058