LAG: Lazily Aggregated Gradient for Communication-Efficient Distributed Learning

TL;DR

This paper introduces LAG, a gradient method that adaptively skips gradient calculations to reduce communication in distributed learning, maintaining convergence rates while saving computational resources.

Contribution

The paper proposes LAG, a novel gradient method that adaptively reuses outdated gradients, achieving communication efficiency without sacrificing convergence guarantees.

Findings

Convergence rate matches standard gradient descent in various settings.

Significant reduction in communication rounds in heterogeneous data scenarios.

Numerical experiments confirm reduced communication with maintained accuracy.

Abstract

This paper presents a new class of gradient methods for distributed machine learning that adaptively skip the gradient calculations to learn with reduced communication and computation. Simple rules are designed to detect slowly-varying gradients and, therefore, trigger the reuse of outdated gradients. The resultant gradient-based algorithms are termed Lazily Aggregated Gradient --- justifying our acronym LAG used henceforth. Theoretically, the merits of this contribution are: i) the convergence rate is the same as batch gradient descent in strongly-convex, convex, and nonconvex smooth cases; and, ii) if the distributed datasets are heterogeneous (quantified by certain measurable constants), the communication rounds needed to achieve a targeted accuracy are reduced thanks to the adaptive reuse of lagged gradients. Numerical experiments on both synthetic and real data corroborate a significant communication reduction compared to alternatives.