Extended ADMM and BCD for nonseparable convex minimization models with quadratic coupling terms: convergence analysis and insights
- 808 Downloads
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 proximal BCD for 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.
KeywordsNonseparable convex minimization Alternating direction method of multipliers Block coordinate descent method Iterate convergence Random permutation
Mathematics Subject Classification65K05 90C26
Caihua Chen was supported by the National Natural Science Foundation of China [Grant No. 11401300, 71732003, 71673130]. Min Li was supported by the National Natural Science Foundation of China [Grant No.11771078, 71390335, 71661147004]. Xin Liu was supported by the National Natural Science Foundation of China [Grant No. 11622112, 11471325, 91530204, 11331012, 11461161005, 11688101], the National Center for Mathematics and Interdisciplinary Sciences, CAS, the Youth Innovation Promotion Association, CAS, and Key Research Program of Frontier Sciences, CAS. Yinyu Ye was supported by the AFOSR Grant [Grant No. FA9550-12-1-0396]. The authors would like to thank Dr. Ji Liu from University of Rochester and Dr. Ruoyu Sun from Stanford University for the helpful discussions on the block coordinate descent method. The authors would also like to thank the associate editor and two anonymous referees for their detailed and valuable comments and suggestions.
- 6.Bertsekas, D.P., Tsitsiklis, J.N.: A dual algorithm for the solution of nonlinear variational problems via finite element approximation. In: Parallel and Distributed Computation: Numerical Methods, 2nd ed. Athena Scientific, Belmont, MA (1997)Google Scholar
- 11.Chen, C., Shen, Y., You, Y.: On the convergence analysis of the alternating direction method of multipliers with three blocks. Abstract and Applied Analysis, 2013, Article ID 183961, 7 pagesGoogle Scholar
- 15.Davis, D., Yin, W.: Convergence rate analysis of several splitting schemes. UCLA CAM Report, 14–51 (2014)Google Scholar
- 19.Feng, C., Xu, H., Li, B.C.: An alternating direction method approach to cloud traffic management. IEEE Trans. Parallel Distrib. Syst. 28(8), 2145–2158 (2017)Google Scholar
- 29.Hong, M., Chang, T., Wang, X., Razaviyayn, M., Ma, S., Luo, Z.: A block successive upper bound minimization method of multipliers for linearly constrained convex optimization. arXiv:1401.7079 (2014)
- 32.Hong, M., Wang, X., Razaviyayn, M., Luo, Z.: Iteration complexity analysis of block coordinate descent methods. arXiv:1310.6957v2 (2014)
- 42.Peng, Y.G., Ganesh, A., Wright, J., Xu, W.L., Ma, Y.: Robust alignment by sparse and low-rank decomposition for linearly correlated images. IEEE Trans. Pattern Anal. Mach. Intell. 34, 2233–2246 (2012)Google Scholar
- 48.Sun, R., Luo, Z., Ye, Y.: On the expected convergence of randomly permuted ADMM. arXiv:1503.06387v1 (2015)
- 52.Zhang Y (2010) Convergence of a class of stationary iterative methods for saddle point problems. Rice University Technique Report, TR10-24Google Scholar