Stochastic-Sign SGD for Federated Learning with Theoretical Guarantees

TL;DR

This paper introduces Stochastic-Sign SGD, a theoretically sound and communication-efficient federated learning algorithm that combines privacy, Byzantine resilience, and heterogeneity handling, validated through extensive experiments.

Contribution

It proposes a novel stochastic-sign based gradient compressor for federated learning, unifying multiple desirable properties in a simple, theoretically justified framework.

Findings

Effective in heterogeneous data settings

Achieves communication efficiency and privacy

Validated on MNIST and CIFAR-10 datasets

Abstract

Federated learning (FL) has emerged as a prominent distributed learning paradigm. FL entails some pressing needs for developing novel parameter estimation approaches with theoretical guarantees of convergence, which are also communication efficient, differentially private and Byzantine resilient in the heterogeneous data distribution settings. Quantization-based SGD solvers have been widely adopted in FL and the recently proposed SIGNSGD with majority vote shows a promising direction. However, no existing methods enjoy all the aforementioned properties. In this paper, we propose an intuitively-simple yet theoretically-sound method based on SIGNSGD to bridge the gap. We present Stochastic-Sign SGD which utilizes novel stochastic-sign based gradient compressors enabling the aforementioned properties in a unified framework. We also present an error-feedback variant of the proposed Stochastic-Sign SGD which further improves the learning performance in FL. We test the proposed method with extensive experiments using deep neural networks on the MNIST dataset and the CIFAR-10 dataset. The experimental results corroborate the effectiveness of the proposed method.