English

Weighted iteration complexity of the sPADMM on the KKT residuals for convex composite optimization

Optimization and Control 2016-11-11 v1

Abstract

In this paper we establish an O(1/k)\mathcal{O}({1}/{k}) weighted iteration complexity on the KKT residuals yielded by the sPADMM (semi-proximal alternating direction method of multiplier) for the convex composite optimization problem. This result, which is derived with the help of a novel generalized HPE (hybrid proximal extra-gradient) iteration formula, first fills the gap on the ergodic iteration complexity of the classic ADMM with a large step-size and its many proximal variants.

Keywords

Cite

@article{arxiv.1611.03167,
  title  = {Weighted iteration complexity of the sPADMM on the KKT residuals for convex composite optimization},
  author = {Li Shen and Shaohua Pan},
  journal= {arXiv preprint arXiv:1611.03167},
  year   = {2016}
}