An Operator Learning Approach to Nonsmooth Optimal Control of Nonlinear PDEs

Explore an innovative operator learning methodology for tackling nonsmooth optimal control problems involving nonlinear partial differential equations in this 49-minute conference talk by Xiaoming Yuan at BIMSA's ICBS2025. Delve into advanced mathematical techniques that bridge machine learning and control theory to address challenging optimization scenarios where traditional smooth optimization methods fall short. Learn how operator learning frameworks can be effectively applied to handle the inherent difficulties of nonsmooth control problems in nonlinear PDE systems, gaining insights into cutting-edge computational approaches that combine deep learning with rigorous mathematical foundations for solving complex control optimization challenges.