A Concentration Bound for Distributed Stochastic Approximation

We revisit the classical model of Tsitsiklis, Bertsekas and Athans [10] for distributed stochastic approximation with consensus. The main result is an analysis of this scheme using the ‘ODE’ (for ‘Ordinary Differential Equations’) approach to stochastic approximation, leading to a high probability bound for the tracking error between suitably interpolated iterates and the limiting differential equation. Several future directions will also be highlighted.