A Communication-Efficient Decentralized Newton's Method with Provably Faster Convergence

TL;DR

This paper introduces a communication-efficient decentralized Newton's method that achieves faster convergence rates than first-order methods by combining dynamic consensus, compression with error correction, and multi-step consensus.

Contribution

It presents the first super-linear convergence analysis for a communication-efficient decentralized Newton's method, improving upon existing first-order approaches.

Findings

Achieves globally linear and super-linear convergence rates.

Uses compression with error compensation for Hessian exchange.

Numerical results confirm theoretical convergence improvements.

Abstract

In this paper, we consider a strongly convex finite-sum minimization problem over a decentralized network and propose a communication-efficient decentralized Newton's method for solving it. We first apply dynamic average consensus (DAC) so that each node is able to use a local gradient approximation and a local Hessian approximation to track the global gradient and Hessian, respectively. Second, since exchanging Hessian approximations is far from communication-efficient, we require the nodes to exchange the compressed ones instead and then apply an error compensation mechanism to correct for the compression noise. Third, we introduce multi-step consensus for exchanging local variables and local gradient approximations to balance between computation and communication. To avoid each node transmitting the entire local Hessian approximation, we design a compression procedure with error compensation to estimate the global Hessian in a communication-efficient way. With novel analysis, we establish the globally linear (resp., asymptotically super-linear) convergence rate of the proposed method when m is constant (resp., tends to infinity), where m is the number of consensus inner steps. To the best of our knowledge, this is the first super-linear convergence result for a communication-efficient decentralized Newton's method. Moreover, the rate we establish is provably faster than those of first-order methods. Our numerical results on various applications corroborate the theoretical findings.