Communication-Efficient Distributed Strongly Convex Stochastic Optimization: Non-Asymptotic Rates

MaRDI QIDQ6306440zbMATH OpenFDO

Authors Anit Kumar Sahu, Dušan Jakovetić, Dragana Bajović, Soummya Kar

Publication date 9 September 2018

Abstract: We examine fundamental tradeoffs in iterative distributed zeroth and first order stochastic optimization in multi-agent networks in terms of emph{communication cost} (number of per-node transmissions) and emph{computational cost}, measured by the number of per-node noisy function (respectively, gradient) evaluations with zeroth order (respectively, first order) methods. Specifically, we develop novel distributed stochastic optimization methods for zeroth and first order strongly convex optimization by utilizing a probabilistic inter-agent communication protocol that increasingly sparsifies communications among agents as time progresses. Under standard assumptions on the cost functions and the noise statistics, we establish with the proposed method the $O(1/(C_{mathrmcomm}{)}^{4/3-zeta})$ and $O(1/(C_{mathrmcomm}{)}^{8/9-zeta})$ mean square error convergence rates, for the first and zeroth order optimization, respectively, where $C_{mathrmcomm}$ is the expected number of network communications and $zeta>0$ is arbitrarily small. The methods are shown to achieve order-optimal convergence rates in terms of computational cost~$C_{mathrmcomp}$, $O(1/C_{mathrmcomp})$ (first order optimization) and $O(1/(C_{mathrmcomp}{)}^{2/3})$ (zeroth order optimization), while achieving the order-optimal convergence rates in terms of iterations. Experiments on real-life datasets illustrate the efficacy of the proposed algorithms.

This page was built for publication: Communication-Efficient Distributed Strongly Convex Stochastic Optimization: Non-Asymptotic Rates

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6306440)