Accelerated dual-averaging primal–dual method for composite convex minimization. (3rd July 2020)

• Record Type: Journal Article
• Authors: Tan, Conghui; Qian, Yuqiu; Ma, Shiqian; Zhang, Tong
• Is Part Of: Optimization methods and software. Volume 35:Number 4(2020)
• Page Start: 741
• Page End: 766
• Publication Date: 2020-07-03
• Subjects: Dual averaging algorithm -- primal–dual -- empirical risk minimization -- acceleration -- sparse data; 90c25; Mathematical optimization -- Periodicals; Algorithms -- Periodicals; 519.7
• DOI: 10.1080/10556788.2020.1713779
• Languages: English
• ISSNs: 1055-6788

Abstract

ABSTRACT: Dual averaging-type methods are widely used in industrial machine learning applications due to their ability to promoting solution structure (e.g. sparsity) efficiently. In this paper, we propose a novel accelerated dual-averaging primal–dual algorithm for minimizing a composite convex function. We also derive a stochastic version of the proposed method that solves empirical risk minimization, and its advantages on handling sparse data are demonstrated both theoretically and empirically.