Stochastic Controlled Averaging for Federated Learning with Communication Compression

TL;DR

This paper introduces two new communication-efficient federated learning algorithms, SCALLION and SCAFCOM, that effectively handle data heterogeneity and compression without stringent assumptions, reducing communication costs while maintaining high performance.

Contribution

It revisits and simplifies stochastic controlled averaging, then develops two novel algorithms supporting unbiased and biased compression with improved efficiency and flexibility.

Findings

Outperform existing compressed FL methods in communication and computation efficiency.

Match full-precision FL performance with significantly reduced uplink communication.

Handle arbitrary data heterogeneity without extra assumptions on compression errors.

Abstract

Communication compression, a technique aiming to reduce the information volume to be transmitted over the air, has gained great interests in Federated Learning (FL) for the potential of alleviating its communication overhead. However, communication compression brings forth new challenges in FL due to the interplay of compression-incurred information distortion and inherent characteristics of FL such as partial participation and data heterogeneity. Despite the recent development, the performance of compressed FL approaches has not been fully exploited. The existing approaches either cannot accommodate arbitrary data heterogeneity or partial participation, or require stringent conditions on compression. In this paper, we revisit the seminal stochastic controlled averaging method by proposing an equivalent but more efficient/simplified formulation with halved uplink communication costs. Building upon this implementation, we propose two compressed FL algorithms, SCALLION and SCAFCOM, to support unbiased and biased compression, respectively. Both the proposed methods outperform the existing compressed FL methods in terms of communication and computation complexities. Moreover, SCALLION and SCAFCOM accommodates arbitrary data heterogeneity and do not make any additional assumptions on compression errors. Experiments show that SCALLION and SCAFCOM can match the performance of corresponding full-precision FL approaches with substantially reduced uplink communication, and outperform recent compressed FL methods under the same communication budget.