AFed: Algorithmic Fair Federated Learning

TL;DR

AFed introduces a novel framework for promoting group fairness in federated learning by learning the global data distribution through generative models, addressing data access restrictions and improving fairness outcomes.

Contribution

The paper proposes AFed, a new approach that uses generative models to enhance fairness in federated learning without accessing local client data.

Findings

Significant fairness improvements over baselines

Effective use of generative models for bias mitigation

Theoretical justification of the methods

Abstract

Federated Learning (FL) has gained significant attention as it facilitates collaborative machine learning among multiple clients without centralizing their data on a server. FL ensures the privacy of participating clients by locally storing their data, which creates new challenges in fairness. Traditional debiasing methods assume centralized access to sensitive information, rendering them impractical for the FL setting. Additionally, FL is more susceptible to fairness issues than centralized machine learning due to the diverse client data sources that may be associated with group information. Therefore, training a fair model in FL without access to client local data is important and challenging. This paper presents AFed, a straightforward yet effective framework for promoting group fairness in FL. The core idea is to circumvent restricted data access by learning the global data distribution. This paper proposes two approaches: AFed-G, which uses a conditional generator trained on the server side, and AFed-GAN, which improves upon AFed-G by training a conditional GAN on the client side. We augment the client data with the generated samples to help remove bias. Our theoretical analysis justifies the proposed methods, and empirical results on multiple real-world datasets demonstrate a substantial improvement in AFed over several baselines.