Federated Learning with Fair Worker Selection: A Multi-Round Submodular Maximization Approach

TL;DR

This paper addresses fair worker selection in federated learning by formulating it as a multi-round submodular maximization problem with fairness constraints, proposing algorithms with theoretical guarantees and validating them through simulations.

Contribution

It introduces a novel multi-round submodular maximization framework with fairness constraints for federated learning, along with three algorithms and theoretical bounds.

Findings

Algorithms achieve fairness and high utility in simulations

FairDG provides stronger short-term fairness guarantees

The proposed methods outperform baseline approaches

Abstract

In this paper, we study the problem of fair worker selection in Federated Learning systems, where fairness serves as an incentive mechanism that encourages more workers to participate in the federation. Considering the achieved training accuracy of the global model as the utility of the selected workers, which is typically a monotone submodular function, we formulate the worker selection problem as a new multi-round monotone submodular maximization problem with cardinality and fairness constraints. The objective is to maximize the time-average utility over multiple rounds subject to an additional fairness requirement that each worker must be selected for a certain fraction of time. While the traditional submodular maximization with a cardinality constraint is already a well-known NP-Hard problem, the fairness constraint in the multi-round setting adds an extra layer of difficulty. To address this novel challenge, we propose three algorithms: Fair Continuous Greedy (FairCG1 and FairCG2) and Fair Discrete Greedy (FairDG), all of which satisfy the fairness requirement whenever feasible. Moreover, we prove nontrivial lower bounds on the achieved time-average utility under FairCG1 and FairCG2. In addition, by giving a higher priority to fairness, FairDG ensures a stronger short-term fairness guarantee, which holds in every round. Finally, we perform extensive simulations to verify the effectiveness of the proposed algorithms in terms of the time-average utility and fairness satisfaction.