1: \begin{abstract}

2:   This paper addresses a class of constrained optimization problems over

3:   networks in which local cost functions and constraints can be nonconvex. We

4:   propose an asynchronous distributed optimization algorithm, relying on the

5:   centralized Method of Multipliers, in which each node wakes up in an

6:   uncoordinated fashion and performs either a descent step on a local Augmented

7:   Lagrangian or an ascent step on the local multiplier vector. These two phases

8:   are regulated by a distributed logic-AND, which allows nodes to understand

9:   when the descent on the (whole) Augmented Lagrangian is sufficiently small. We

10:   show that this distributed algorithm is equivalent to a block coordinate

11:   descent algorithm for the minimization of the Augmented

12:   Lagrangian followed by an update of the whole multiplier vector. Thus, the proposed

13:   algorithm inherits the convergence properties of the Method of Multipliers.

14: %  To the best of our knowledge this is the first asynchronous fully distributed

15: %  algorithm for the addressed class of problems.

16: \end{abstract}

17: