CONVERGENCE RATES OF PROXIMAL GRADIENT METHODS VIA THE CONVEX CONJUGATE

被引:2
|
作者
Gutman, David H. [1 ]
Pena, Javier F. [2 ]
机构
[1] Carnegie Mellon Univ, Dept Math Sci, Pittsburgh, PA 15213 USA
[2] Carnegie Mellon Univ, Tepper Sch Business, Pittsburgh, PA 15213 USA
关键词
convex conjugate; proximal gradient; acceleration; ALGORITHMS;
D O I
10.1137/18M1164329
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We give a novel proof of the O(1/k) and O(1/k(2)) convergence rates of the proximal gradient and accelerated proximal gradient methods for composite convex minimization. The crux of the new proof is an upper bound constructed via the convex conjugate of the objective function.
引用
收藏
页码:162 / 174
页数:13
相关论文
共 50 条