Type: Seminar

Modified: Thu 3 Oct 2024 1511; Thu 3 Oct 2024 1518

Distribution: World

Expiry: 16 Oct 2024

CalTitle1: SMRI Seminar: Johannes Wiesel TBA

Auth: cmeister@w95s10l3.staff.sydney.edu.au (cmei0631) in SMS-SAML

Speaker: Johannes Wiesel Title: Estimating processes using optimal transport Abstract: The Wasserstein distance $\mathcal{W}_p$ is an important instance of an optimal transport cost. Its numerous mathematical properties have been well studied in recent years. The adapted Wasserstein distance $\mathcal{AW}_p$ extends this theory to laws of discrete time stochastic processes in their natural filtrations, making it particularly well suited for analyzing time-dependent stochastic optimization problems. Recently, $\mathcal{W}_p$ has found a lot of interesting applications in statistics and machine learning. In this talk I will explain how some of these results can be extended to stochastic processes and $\mathcal{AW}_p$. In particular I will focus a new measure of dependence between two random objects, which I call the Wasserstein correlation.