Speaker : Dohyun Kwon(University of seoul) 

Abstract: The study of gradient flows holds significant importance across various fields, including partial differential equations, optimization, and machine learning. In this talk, we aim to explore the relationship between gradient flows and their time-discretized formulations, known as De Giorgi's minimizing movements scheme. We focus on how De Giorgi's minimizing movements coincide with gradient flows in two different spaces: the space of sets and the space of probability measures called Wasserstein space. Then, we discuss their implications for the well-posedness and long-time behavior of some PDEs, including mean curvature flow and the nonlinear Fokker-Planck equation.