Local Linear Convergence of a Primal-Dual Algorithm for the Augmented Convex Models

被引：0
作者：
Tao Sun
Roberto Barrio
Hao Jiang
Lizhi Cheng
机构：
[1] National University of Defense Technology,College of Science
[2] Universidad de Zaragoza,Departamento de Matemática Aplicada, IUMA and CoDy group
[3] National University of Defense Technology,College of Computer
[4] National University of Defense Technology,College of Science and The State Key Laboratory for High Performance Computation
来源：
Journal of Scientific Computing | 2016年 / 69卷
关键词：
Convex optimization; Primal-dual algorithm; Augmented convex model; Partial smoothness; Local linear convergence; 90C25; 65K10;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
Convex optimization has become an essential technique in many different disciplines. In this paper, we consider the primal-dual algorithm minimizing augmented models with linear constraints, where the objective function consists of two proper closed convex functions; one is the square of a norm and the other one is a gauge function being partly smooth relatively to an active manifold. Examples of this situation can be found in signal processing, optimization, statistics and machine learning literature. We present a unified framework to understand the local convergence behaviour of the primal-dual algorithm for these augmented models. This result explains some numerical results shown in the literature on local linear convergence of the algorithm.
引用
收藏
页码：1301 / 1315
页数：14
相关论文
共 48 条
  • [1] Chen SS(1998)Atomic decomposition by basis pursuit SIAM J. Sci. Comput. 20 33-61
  • [2] Donoho DL(2007)Analysis versus synthesis in signal priors Inverse Probl. 23 947-1350
  • [3] Saunders MA(2008)Exact regularization of convex programs SIAM J. Optim. 18 1326-1130
  • [4] Elad M(2008)Fixed-point continuation for SIAM J. Optim. 19 1107-266
  • [5] Milanfar P(2004)-minimization: methodology and convergence J. Conv. Anal. 11 251-453
  • [6] Rubinstein R(2013)Identifying active constraints via partial smoothness and prox-regularity J. Sci. Comput. 54 428-1091
  • [7] Friedlander MP(2013)Accelerated linearized Bregman method SIAM J. Imaging Sci. 6 1059-725
  • [8] Tseng P(2002)Augmented SIAM J. Optim. 13 702-234
  • [9] Hale ET(2008) and nuclear-norm models with a globally linearly convergent algorithm Math. Oper. Res. 33 216-426
  • [10] Yin W(2012)Active sets, nonsmoothness, and sensitivity Inverse Probl. 28 095003-111
12345