1: \begin{abstract}

2: We propose a novel algorithmic framework of {\bf V}ariable {\bf M}etric {\bf O}ver-{\bf R}elaxed  {\bf H}ybrid {\bf P}roximal {\bf E}xtra-gradient (VMOR-HPE) method with a global convergence guarantee for the maximal monotone operator inclusion problem.

3: Its iteration complexities and local linear convergence rate are provided, which theoretically demonstrate that a large over-relaxed step-size contributes to accelerating the proposed VMOR-HPE as a byproduct. Specifically, we find that a large class of primal and primal-dual operator splitting algorithms are all special cases of VMOR-HPE. Hence, the proposed framework offers a new insight into these operator splitting algorithms.

4: In addition, we apply VMOR-HPE to the Karush-Kuhn-Tucker (KKT) generalized equation of linear equality constrained multi-block composite convex optimization, yielding a new algorithm, namely nonsymmetric {\bf P}roximal {\bf A}lternating {\bf D}irection {\bf M}ethod of {\bf M}ultipliers with a preconditioned {\bf E}xtra-gradient step in which the preconditioned metric is generated by a blockwise {\bf B}arzilai-{\bf B}orwein line search technique (PADMM-EBB).

5: We also establish iteration complexities of PADMM-EBB in terms of the KKT residual.

6: Finally, we apply PADMM-EBB to handle the nonnegative dual graph regularized low-rank representation problem.

7: Promising results on synthetic and real datasets corroborate the efficacy of PADMM-EBB.

8: \vspace{-13pt}

9: \end{abstract}

10: