1: \begin{abstract}

2:  In this paper, a distributed stochastic approximation algorithm is

3: studied.  Applications of such algorithms include decentralized

4: estimation, optimization, control or computing. The algorithm consists

5: in two steps: a local step, where each node in a network updates a

6: local estimate using a stochastic approximation algorithm with

7: decreasing step size, and a gossip step, where a node computes a local

8: weighted average between its estimates and those of its

9: neighbors. Convergence of the estimates toward a consensus is

10: established under weak assumptions. The approach relies on two main

11: ingredients: the existence of a Lyapunov function for the mean field

12: in the agreement subspace, and a contraction property of the random

13: matrices of weights in the subspace orthogonal to the agreement

14: subspace.  A second order analysis of the algorithm is also performed

15: under the form of a Central Limit Theorem. The Polyak-averaged version

16: of the algorithm is also considered.

17: \end{abstract}