Extended ADMM and BCD for nonseparable convex minimization models with quadratic coupling terms: convergence analysis and insights

Chen Caihua Nanjing University Li Min Nanjing University Liu Xin Chinese Academy of Sciences and University of Chinese Academy of Sciences Ye Yinyu Stanford University

Optimization and Control mathscidoc:1803.27002

Mathematical Programming, 2017.10
In this paper,we establish the convergence of the proximal alternating direction method of multipliers (ADMM) and block coordinate descent (BCD) method for nonseparable minimization models with quadratic coupling terms. The novel convergence results presented in this paper answer several open questions that have been the subject of considerable discussion. We firstly extend the 2-block proximal ADMM to linearly constrained convex optimization with a coupled quadratic objective function, an area where theoretical understanding is currently lacking, and prove that the sequence generated by the proximal ADMM converges in point-wise manner to a primal-dual solution pair. Moreover, we apply randomly permuted ADMM (RPADMM) to nonseparable multi-block convex optimization, and prove its expected convergence for a class of nonseparable quadratic programming problems. When the linear constraint vanishes, the 2-block proximal ADMM and RPADMM reduce to the 2-block cyclic proximal BCD method and randomly permuted BCD (RPBCD). Our study provides the first iterate convergence result for 2-block cyclic proximal BCD without assuming the boundedness of the iterates. We also theoretically establish the expected iterate convergence result concerning multi-block RPBCD for convex quadratic optimization. In addition, we demonstrate that RPBCD may have a worse convergence rate than cyclic proximalBCDfor 2-block convex quadratic minimization problems. Although the results on RPADMM and RPBCD are restricted to quadratic minimization models, they provide some interesting insights: (1) random permutation makes ADMM and BCD more robust for multi-block convex minimization problems; (2) cyclic BCD may outperform RPBCD for “nice” problems, and RPBCD should be applied with caution when solving general convex optimization problems especially with a few blocks.
Nonseparable convex minimization, Alternating direction method of multipliers, Block coordinate descent method, Iterate convergence, Random permutation
[ Download ] [ 2018-03-04 17:11:40 uploaded by chchen ] [ 532 downloads ] [ 0 comments ]
@inproceedings{chen2017extended,
  title={Extended ADMM and BCD for nonseparable convex minimization models with quadratic coupling terms: convergence analysis and insights},
  author={Chen Caihua, Li Min, Liu Xin, and Ye Yinyu},
  url={http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20180304171140958616965},
  booktitle={Mathematical Programming},
  year={2017},
}
Chen Caihua, Li Min, Liu Xin, and Ye Yinyu. Extended ADMM and BCD for nonseparable convex minimization models with quadratic coupling terms: convergence analysis and insights. 2017. In Mathematical Programming. http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20180304171140958616965.
Please log in for comment!
 
 
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved