报告题目:Difference of convex algorithm for bilevel programs with applications in hyperparameter selection
报 告人: 曾尚志 博士
报告时间: 2023年3月12 日(星期日)10:20-11:05
报告地点:四号楼4318会议室
邀 请人: 潘少华
欢迎广大师生前往!
数学学院
2023年3月8日
报告摘要:In this work, we present a difference of convex algorithm for solving bilevel programs in which the upper level objective functions are difference of convex functions, and the lower level programs are fully convex. This nontrivial class of bilevel programs provides a powerful modelling framework for dealing with applications arising from hyperparameter selection in machine learning. Thanks to the full convexity of the lower-level program, the value function of the lower level program turns out to be convex and hence the bilevel program can be reformulated as a difference of convex bilevel program. We propose an algorithm for solving the reformulated difference of convex program and show its convergence to stationary points under very mild assumptions.
报告人简介:加拿大维多利亚大学PIMS博士后。本科毕业于武汉大学,硕士毕业于香港浸会大学,2021年博士毕业于香港大学。主要研究方向是变分分析及双层规划,成果发表在Math Program (2篇)、SIAM J Numer Anal、J Mach Learn Res、IEEE T Pattern Anal Mach Intell、IEEE T Image Process,以及ICML (4篇)、NeurIP (spotlight)等优化、计算数学、机器学习期刊与会议上。