报告地点：腾讯会议：211-577-881
邀请人：孙海琳教授
报告摘要：
It has been widely recognized that the cardinality function is one of the most natural choices for modelling classification errors, and it has a wide range of applications including support vector machines and 1-bit compressed sensing. Due to the combinatorial nature of the cardinality function, methods based on convex relaxations or smoothing approximations have dominated the existing research and are often able to provide approximate solutions of good quality. However, those methods are not optimizing the cardinality function directly and hence no optimality has been established for the original problem. This talk aims to study the optimality conditions of the cardinality function minimization and for the first time to develop Newton's method that directly optimizes the cardinality function with a local quadratic convergence under reasonable conditions.
报告人简介：潘丽丽 山东理工大学副教授，2017年获得北京交通大学博士学位，多次访问香港理工大学。主持国家自然科学基金青年项目一项，在《SIAM Journal on Imaging Sciences》、《SIAM Journal on Optimization》、《Numerical Algorithms》等期刊发表SCI论文多篇。