Adaptive Stochastic Algorithms for Nonconvex Constrained Optimization

In this talk, we will discuss some recent works on the design, analysis, and implementation of a class of efficient algorithms for solving stochastic optimization problems with deterministic nonlinear nonconvex constraints. Those optimization problems arise in a plethora of science and engineering applications including physics-informed learning, PDE-constrained optimization, machine learning fairness, and optimal power flow. We are especially interested in the case where the problem's feasible region is difficult to detect and projection-type methods are intractable. The theoretical results and numerical performance demonstrate the efficiency and efficacy of our proposed algorithms.

Bio
https://baoyuzhou18.github.io/