1: \begin{abstract}

2: We present a so-called universal convergence theorem for inexact primal-dual penalty and augmented Lagrangian methods

3: that can be applied to a large number of such methods and reduces their convergence analysis to verification of some 

4: simple conditions on sequences generated by these methods. If these conditions are verified, then both primal and dual

5: convergence follow directly from the universal convergence theorem. This theorem allows one not only to derive standard

6: convergence theorems for many existing primal-dual penalty and augmented Lagrangian methods in a unified and

7: straightforward manner, but also to strengthen and generalize some of these theorems. In particular, we show how with

8: the use of the universal convergence theorem one can significantly improve some existing results on convergence of 

9: a primal-dual rounded weighted $\ell_1$-penalty method, an augmented Lagrangian method for cone constrained

10: optimization, and some other primal-dual methods.

11: \end{abstract}

12: