报告人：孙海琳 教授（南京师范大学）

时间：2024年12月05日 16:00-

地点: 数统学院LD402

摘要：Stochastic approximation (SA) type methods have been well studied for solving single-stage stochastic variational inequalities (SVIs). In this talk, we consider SA type methods for two-stage SVIs and multistage SVIs. In particular, we propose a dynamic stochastic projection method (DSPM) for solving multistage SVIs under strongly monotone conditions. For non-strongly monotone two-stage SVIs, we propose dynamic sampling stochastic projection gradient method (DS-SPGM) for solving a class of two-stage SVIs with co-coercive property. The corresponding convergence rates are investigated. Numerical experiments show the efficiency of the two algorithms.

邀请人: 蒋杰

欢迎广大师生积极参与！