报告题目:TowardsUnderstanding the Convergence Behavior of Newton-Type Methods forStructured Convex Optimization Problems
报告人:苏文藻教授(香港中文大学)
报告时间:2017年03月16日(星期四)下午15:00-17:00
报告地点:4号楼4318室
欢迎广大师生前往!
数学学院
2017年03月13日
报告摘要:Recently,there has been a growing interest in applying Newton-type methods tosolve structured
convexoptimization problems that arise in machine learning and statistics.A major obstacle to the design
andanalysis of such methods is that many problems of interest areneither strongly convex nor smooth. In this
talk, wewill present some design techniques for overcoming such obstacle andreport some recent progress
onanalyzing the convergence rates of the resulting Newton-type methodsusing error bounds. We will also
discusssome directions for further study.