Towards Collaborative Fairness in Federated Learning Under Imbalanced Covariate Shift

TL;DR

This paper introduces FedAKD, a federated learning method that improves collaborative fairness and accuracy under imbalanced covariate shift by using asynchronous knowledge distillation and selective model updates.

Contribution

The paper presents a novel federated learning algorithm, FedAKD, with theoretical convergence proof and practical effectiveness in handling heterogeneity and imbalanced data distributions.

Findings

FedAKD improves fairness and accuracy on multiple datasets.

The approach enhances client participation under heterogeneity.

Theoretical convergence of FedAKD is established.

Abstract

Collaborative fairness is a crucial challenge in federated learning. However, existing approaches often overlook a practical yet complex form of heterogeneity: imbalanced covariate shift. We provide a theoretical analysis of this setting, which motivates the design of FedAKD (Federated Asynchronous Knowledge Distillation)- simple yet effective approach that balances accurate prediction with collaborative fairness. FedAKD consists of client and server updates. In the client update, we introduce a novel asynchronous knowledge distillation strategy based on our preliminary analysis, which reveals that while correctly predicted samples exhibit similar feature distributions across clients, incorrectly predicted samples show significant variability. This suggests that imbalanced covariate shift primarily arises from misclassified samples. Leveraging this insight, our approach first applies traditional knowledge distillation to update client models while keeping the global model fixed. Next, we select correctly predicted high-confidence samples and update the global model using these samples while keeping client models fixed. The server update simply aggregates all client models. We further provide a theoretical proof of FedAKD's convergence. Experimental results on public datasets (FashionMNIST and CIFAR10) and a real-world Electronic Health Records (EHR) dataset demonstrate that FedAKD significantly improves collaborative fairness, enhances predictive accuracy, and fosters client participation even under highly heterogeneous data distributions.