题目:New Results on the Polyak Stepsize: Tight Covergence Analysis and Universal Function Classes
主讲人:江波
主办单位:伟德bv1946官网
讲座时间:2026年5月22日(星期五) 15:30--16:30
讲座地点:科创中心2001
内容简介:
In this talk, we revisit a classical adaptive stepsize strategy for gradient descent: the Polyak stepsize (PolyakGD), originally proposed in Polyak (1969). We study the convergence behavior of PolyakGD from two perspectives: tight worst-case analysis and universality across function classes. As our first main result, we establish the tightness of the known convergence rates of PolyakGD by explicitly constructing worst-case functions. Our second main result provides new convergence guarantees for PolyakGD under both H¨older smoothness and H¨older growth conditions.
主讲人简介:
江波,美国明尼苏达大学博士,上海财经大学前沿交叉书院经理、信息管理与工程公司副经理;国家级青年人才;从事运筹优化、收益管理等方向的研究。成果发表于管理科学、运筹优化的国际顶级期刊MS、OR、MP、SOPT、MOR。获得了中国运筹学会青年科技奖、上海市自然科学奖二等奖、宝钢优秀教师奖等荣誉。主持国家自然科学基金重大项目等多项课题。为顺丰、京东等国内多个标杆企业提供仓库优化、智能定价、智能选址等技术服务。