学术报告

题      目： Proximal and Splitting Algorithms in Optimization

报  告  人：杨俊锋   教授  (邀请人：陈艳男 )

                                南京大学

时      间：2022-11-04 15:00-16:00

腾 讯 会 议：454 239 227

报告人简介:

       杨俊锋，南京大学数学系教授，博导，主要从事最优化计算方法及其应用研究，在SIAM系列、Mathematics of Computation等杂志上发表研究论文多篇，开发图像去模糊软件包FTVd, 压缩感知一模解码软件包YALL1, 核磁共振图像复原软件包RecPF等。2012 年入选教育部新世纪优秀人才支持计划，2016年获中国运筹学会青年科技奖, 主持国家自然科学基金青年、面上、优青项目等多项。

摘      要：

       In this talk, I will first introduce the problems we are interested in, including convex optimization problems, often with separable structures, mixed variational inequality problems and monotone operator inclusion problems. Optimality conditions, examples and various connections in between them will be included. Then, I will talk about proximal point algorithm, augmented Lagrangian method, ADMM, primal-dual algorithms, Douglas-Rachford and Davis-Yin operator splitting algorithms, and the consensus technique. Finally, I will introduce some recent results on frugal resolvent splitting with minimal lifting.