Rate of Convergence for Distributed Optimization with Uncertain Communications
We consider the distributed optimization problem for the sum of convex functions where the underlying communications network connecting agents at each time is drawn at random from a collection of directed graphs. Building on an earlier work [15], where a modified version of the subgradient-push algorithm is shown to be almost surely convergent to an optimizer on sequences of random directed graphs, we find an upper bound of the order of $\sim O(\frac{1}{\sqrt{t}})$ on the convergence rate of our proposed algorithm, establishing the first convergence bound in such random settings.