地点：行健楼学术活动室665
邀请人：孙海琳教授
摘要：This talk considers a class of nonconvex-nonconcave min-max optimization problems with cardinality penalties. We show the existence of local saddle points and global(local) minimax points of the problem. We study the continuous relaxation problem and establish its relationship with the original problem regarding local saddle points and global (local) minimax points. Moreover, we present the optimality conditions for the continuous relaxation problem.A smoothing quasi-Newton subspace trust region algorithm with global convergence is presented for solving nonconvex-nonconcave min-max optimization problems. Preliminary numerical results illustrate the efficiency of the algorithm.
个人简介：
Xiaojun Chen is a Chair Professor of Department of Applied Mathematics, Hong Kong Polytechnic University. She is the Co-Director of CAS AMSS-PolyU Joint Laboratory of Applied Mathematics. Her research interests focus on mathematical optimization theory and algorithms for nonsmooth nonconvex optimization problems and stochastic variational inequalities with applications in data sciences. She is the PI of several large grants from Hong Kong Research Grant Council and Croucher Foundation. She published over 90 papers in top journals in applied mathematics. She is an Associate Editor of SIAM J. Optimization, SIAM J. Numerical Analysis, SIAM J. Control and Optimization, and the Area Editor of Journal on Optimization Theory and Applications. She is a fellow of Society for Industrial and Applied Mathematics and a fellow of American Mathematical Society. She is a Keynote speaker of the 25th International Symposium on Mathematical Programming in Canada 2024.
