Stability and Generalization for Randomized Coordinate Descent

被引：0

作者：

Wang, Puyu ^{[1
]}

Wu, Liang ^{[2
]}

Lei, Yunwen ^{[3
]}

机构：

[1] Northwest Univ, Sch Math, Xian 710127, Peoples R China

[2] Southwestern Univ Finance & Econ, Sch Stat, Ctr Stat Res, Chengdu 611130, Peoples R China

[3] Univ Birmingham, Sch Comp Sci, Birmingham B15 2TT, England

来源：

PROCEEDINGS OF THE THIRTIETH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2021 | 2021年

基金：

中国国家自然科学基金;

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Randomized coordinate descent (RCD) is a popular optimization algorithm with wide applications in solving various machine learning problems, which motivates a lot of theoretical analysis on its convergence behavior. As a comparison, there is no work studying how the models trained by RCD would generalize to test examples. In this paper, we initialize the generalization analysis of RCD by leveraging the powerful tool of algorithmic stability. We establish argument stability bounds of RCD for both convex and strongly convex objectives, from which we develop optimal generalization bounds by showing how to early-stop the algorithm to tradeoff the estimation and optimization. Our analysis shows that RCD enjoys better stability as compared to stochastic gradient descent.

引用

页码：3104 / 3110

页数：7