Distributed Randomized Block Stochastic Gradient Tracking Method

TL;DR

This paper introduces a novel randomized block-coordinate distributed stochastic gradient tracking method designed for large-scale, networked optimization problems, achieving fast convergence rates and validated on real and synthetic datasets.

Contribution

It presents the first randomized block-coordinate gradient tracking algorithm with proven convergence rates for distributed optimization over networks.

Findings

Achieves $1/k$ and $1/k^2$ convergence rates in non-asymptotic analysis.

Demonstrates effectiveness on MNIST and synthetic datasets.

First to incorporate randomized block-coordinate updates in distributed gradient tracking.

Abstract

We consider distributed optimization over networks where each agent is associated with a smooth and strongly convex local objective function. We assume that the agents only have access to unbiased estimators of the gradient of their objective functions. Motivated by big data applications, our goal lies in addressing this problem when the dimensionality of the solution space is possibly large and consequently, the computation of the local gradient mappings may become expensive. We develop a randomized block-coordinate variant of the recently developed distributed stochastic gradient tracking (DSGT) method. We derive non-asymptotic convergence rates of the order $1/k$ and $1/k^2$ in terms of an optimality metric and a consensus violation metric, respectively. Importantly, while block-coordinate schemes have been studied for distributed optimization problems before, the proposed algorithm appears to be the first randomized block-coordinate gradient tracking method that is equipped with the aforementioned convergence rate statements. We validate the performance of the proposed method on the MNIST and a synthetic data set under different network settings.