Distinguished Lecture by Prof. Charles Bouveyron (Inria Centre at Université Côte d’Azur) "Scaling Optimal Transport to High-Dimensional Gaussian Distributions with Application to Domain Adaptation"

A RIKEN AIP Distinguished Lecture by Prof. Charles Bouveyron (Inria Centre at Université Côte d'Azur), held in hybrid form online and at the University of Tokyo. The talk addresses optimal transport (OT) for high-dimensional data, where common approaches project onto low-dimensional spaces and lose information. For high-dimensional Gaussians with parsimonious covariance and low intrinsic dimension, he presents a closed-form 2-Wasserstein distance and Monge map computed without projections, competitive even on non-Gaussian data, with applications to unsupervised domain adaptation. Prof. Bouveyron heads the INRIA MAASAI team and holds an AI chair at Institut 3IA Côte d'Azur.