科学研究

学术活动

New Classes of Non-monotone Variational Inequality Problems Solvable via Proximal Gradient on Smooth Gap Functions
发布时间:2025-10-28 23:37 来源:数学与人工智能学院 点击率:

主讲人:赵磊

主办单位:数学与人工智能学院

讲座时间:2025年10月30日下午14:30-15:30

讲座形式:创新中心2001或腾讯会议841841649

内容简介:

In this talk, we discuss the local linear convergence behavior of proximal-gradient descent algorithm on a parameterized gap-function reformulation of a smooth but non-monotone variational inequality problem. The aim is to solve the non-monotone VI problem without assuming the existence of a Minty-type solution. We first introduce and study various error bound conditions for the gap functions in relation to the VI model. As a result, local linear convergence is established under some easy-verifiable conditions induced by level-set error bounds, the gradient Lipschitz condition and a suitable initialization condition. Furthermore, for non-monotone affine VIs we present a homotopy continuation scheme that achieves global convergence by dynamically tracing a solution path.  

主讲人简介:

赵磊,博士,2019年毕业于上海交适大学安泰经济与管理学院。现为上海交通大学转化医学研究院助理研究员,上海人工智能研究院特聘研究员。主要研究方向为最优化与人工智能、AIforMIedicine。在人工智能会议ICML、Neurips、应用数学MOR、Nature 正刊1 篇,投权国家发明专利3 项。参与研发国产首台套锥形束CT:研发的装箱优化、零部件取件优化、整车配载优化等软件在上汽通用、长城汽车取得了良好的应用效果。在精准放疗计划编制、脉冲电场消融治疗计划与引导等方面取得了一系列成果,部分已形成软件,应用于临床。