FedShuffle: Recipes for Better Use of Local Work in Federated Learning

TL;DR

FedShuffle introduces a novel federated learning method that effectively utilizes local updates with theoretical guarantees, addressing data heterogeneity and random reshuffling to improve convergence and performance.

Contribution

It is the first local update method with convergence guarantees that incorporates random reshuffling, data imbalance, and client sampling in federated learning.

Findings

FedShuffle converges under heterogeneous data and client sampling.

It outperforms FedAvg and FedNova in heterogeneous settings.

Momentum variance reduction enhances FedShuffle's performance.

Abstract

The practice of applying several local updates before aggregation across clients has been empirically shown to be a successful approach to overcoming the communication bottleneck in Federated Learning (FL). Such methods are usually implemented by having clients perform one or more epochs of local training per round while randomly reshuffling their finite dataset in each epoch. Data imbalance, where clients have different numbers of local training samples, is ubiquitous in FL applications, resulting in different clients performing different numbers of local updates in each round. In this work, we propose a general recipe, FedShuffle, that better utilizes the local updates in FL, especially in this regime encompassing random reshuffling and heterogeneity. FedShuffle is the first local update method with theoretical convergence guarantees that incorporates random reshuffling, data imbalance, and client sampling - features that are essential in large-scale cross-device FL. We present a comprehensive theoretical analysis of FedShuffle and show, both theoretically and empirically, that it does not suffer from the objective function mismatch that is present in FL methods that assume homogeneous updates in heterogeneous FL setups, such as FedAvg (McMahan et al., 2017). In addition, by combining the ingredients above, FedShuffle improves upon FedNova (Wang et al., 2020), which was previously proposed to solve this mismatch. Similar to Mime (Karimireddy et al., 2020), we show that FedShuffle with momentum variance reduction (Cutkosky & Orabona, 2019) improves upon non-local methods under a Hessian similarity assumption.