报告人：张雷洪（苏州大学）

时间：2022年06月30日10:00-

腾讯会议ID：227 201 254

摘要：The multi-view canonical correlation analysis is an extension of the canonical correlation analysis (CCA) that is able to deal with multi-view data in many modern applications. Traditional CCA imposes constraint which first whitens the data in the low-dimensional space and then seeks the maximal correlations. However, such a whitening procedure eliminates the data structure in the high dimensional space and limits its application in the sub-sequential learning tasks.  

In this talk, we target the orthogonal multi-view canonical correlation analysis in the form of trace-fractional structure and orthogonal linear projections. We shall discuss properties of the resulting optimization and also propose efficient algorithms. Experiments on real-world applications of multi-label classification and multi-view feature extraction will be reported to demonstrate the efficiency of the proposed method.

简介：张雷洪于2008年博士毕业于香港浸会大学，现为苏州大学数学科学学院教授。研究方向为最优化理论与计算、数值线性代数、数据科学等。主持多项自然青年/面上项目，参与国家自然科学基金重大研究计划。在SIOPT、SIMAX、SISC、Math Comput、Numer Math、IEEE TPMAI等期刊已发表50多篇学术论文。曾获第四届中国数学会计算数学分会 “应用数值代数奖’’、上海财经大学第四届学术奖、2018和2019年世界华人数学家联盟最佳论文奖（若琳奖），及2019年上海市自然科学三等奖（第一完成人） 等。现为学术杂志《Operators and Matrices》和《Numerical Algebra, Control and Optimization》的编委。

邀请人：李寒宇

欢迎广大师生积极参与！