DMS Applied and Computational Mathematics Seminar

Time: Nov 04, 2022 (02:00 PM)
Location: ZOOM



Speaker:  Rongjie Lai, Rensselaer Polytechnic Institute

Title: Computational Methods for Mean-field Games: From Conventional Numerical Methods to Deep Generative Models
Abstract: Mean field game (MFG) problems study how a large number of similar rational agents make strategic movements to minimize their costs. In this talk, I will discuss our recent work on computational methods for MFGs. I will start from a low-dimensional setting using conventional discretization/optimization methods and discuss some convergence results of the proposed method. Then, I will discuss its extension to computing problems on low-dimensional manifolds. After that, I will extend my discussion to high-dimensional problems by bridging the trajectory representation of MFG with a special type of deep generative model, normalizing flows. This connection does not only help solve high-dimensional MFGs, but also provides a way to improve the robustness of normalizing flows. 
Faculty host: Yimin Zhong