1: \begin{abstract}

2: In this paper, we consider the linearly constrained composite convex optimization problem, whose objective is a sum of a smooth function and a possibly nonsmooth function. We propose an inexact augmented Lagrangian (IAL) framework for solving the problem. The proposed IAL framework requires solving the augmented Lagrangian (AL) subproblem at each iteration less accurately than most of the existing IAL frameworks/methods. {We analyze the global convergence and the non-ergodic convergence rate of the proposed IAL framework.}

3: 

4: %The required solution accuracy of the AL subproblem is less than that in most of the existing IAL methods.

5: 

6: %The required stopping condition of solving the AL subproblem in the proposed IAL method is weaker than the one in most of the existing IAL methods.

7: 

8: %At each iteration, the proposed IAL method approximately solves the augmented Lagrangian (AL) subproblem {(with fixed Lagrange multiplier)} within a preselected tolerance and then updates the Lagrange multiplier.

9: % is $O(\epsilon^{-4}).$

10: % we show that for any fixed $0<q<1$, finding the global minimizer of the problem, even its unconstrained counterpart, is strongly NP-hard. Secondly,

11: 

12: %with quadratic programming (QP) oracle and a constant reduction in the smoothing objective is guaranteed

13: %Insert your abstract here. Include keywords, PACS and mathematical

14: %subject classification numbers as needed.

15: \keywords{Inexact augmented Lagrangian framework \and Non-ergodic convergence rate}

16: % \PACS{PACS code1 \and PACS code2 \and more}

17:  \subclass{90C25 \and 65K05}

18: \end{abstract}