Abstract:
In this talk, we explore two areas of modern statistical analysis. First, we examineoverparameterized statistical learning problems, which are central to modern machinelearning. We illustrate how results on random matrices can be applied to prove recentbreakthrough results concerning the benign effects of overparameterization in asimple setting.
In the second part, we turn to statistical challenges in applied optimal transport, witha focus on applications in generative models and barycenters. We will discuss bothrecent developments and ongoing challenges.
This talk is intended to be accessible to a broad audience, with no specializedbackground required.