报告题目：Sparse least squares solutions of tensor equations
报告人：黄正海教授 天津大学
报告时间：2023年11月30日（周四）下午 16:00
报告地点：工科4-638
主办单位：数学与统计学院
报告摘要：
In this talk, we investigate the problem of finding a sparse least squares solution to tensor equations, where the tensor involved needs not be a square tensor. When the tensor involved is reduced to a square tensor, the proposed model is reduced to a model recently studied in the literature. By splitting the tensor involved, we propose a linearized method for solving the problem under consideration, which is an extension of the natural thresholding algorithm proposed recently by Zhao and Luo for solving sparse linear least squares problem. Under proper assumptions, the convergence of the proposed algorithm is proved. Preliminary numerical results show that the algorithm is effective.
报告人简介：
黄正海，天津大学数学学院教授、博士生导师。主要从事最优化理论、算法及其应用方面的研究工作，在求解互补与变分不等式问题、对称锥优化与对称锥互补问题、稀疏优化、张量优化、核磁共振医学成像、人脸识别等方面取得了一系列有意义的成果。目前的主要研究兴趣是张量优化、特殊结构的变分不等式与互补问题、以及机器学习中的优化理论方法及其应用。已发表SCI检索论文130多篇、连续获得多项国家自然科学基金资助。曾获得中科院优秀博士后奖和教育部高等学校自然科学奖二等奖。目前为中国运筹学会常务理事；国际期刊《Pacific Journal of Optimization》、《Asia-Pacific Journal of Operational Research》、《Applied Mathematics and Computation》和《Optimization,Statistics & Information Computing》的编委、中国核心期刊《运筹学学报》的编委。
欢迎广大师生踊跃参加！
