报告题目:Decomposition Methods for Large-ScaleOptimization Problems and Their Applications
报告人:陈彩华
时间:2017年04月25日晚上8:00
地点:汇贤楼122教室
主办单位:十大网投正规信誉官网
摘要:
We live in the age of big data and dataof huge size is becoming ubiquitous. With thiscomesthe need to solve optimization problems of unprecedented size. Classicaloptimization algorithms are not designed to scale to instances of this size. Inthis talk, I introduce two typical decomposition methods —— BCDand ADMM, for solving large scale optimization problems and present somenovel theoretical results on these two methods. Some interestingapplications, including matrix completion, pricing discrimination forinformation goods and sparse portfolio selection are also discussed.
个人简介:
陈彩华,副教授,南京大学理学博士,新加坡国立大学联合培养博士。曾赴新加坡国立大学、香港理工大学、香港浸会大学等学习与访问,与来自新加坡、香港、美国的国际一流学者建立了实质性的合作关系。主要研究方向: 最优化理论、算法与应用高维数据分析计算管理科学/金融。目前主持国家自然科学基金和江苏省自然科学基金各一项,已在《MathematicalProgramming》,《SIAM Journal on Optimization》,《SIAM Journal on Imaging Science》、《IMA Journal of Numerical Analysis》等国际知名学术期刊发表文章十余篇,其中ESI高被引论文2篇。