Features You Always Wanted to Know About the Geometry in Wasserstein Space

par Kexin Lin

lundi 8 déc. 2025, 10:30 → 11:25 Europe/Paris

Salle 112 (ICJ)

Optimal transport allows us to define distances between probability distributions, known as Wasserstein distances. These distances are widely used in applied mathematics particularly in image processing thanks to the robust and geometric properties hidden in these distances.

In this talk, we will begin by introducing the Optimal Transport problem. Next, we will discuss Wasserstein distances and the geodesic curves in Wasserstein space, which naturally arise from the optimal transport map. Finally, through several examples, we will illustrate how the Wasserstein distance reflects the underlying geometry of the distributions.