Navigating Heterogeneity and Privacy in One-Shot Federated Learning with Diffusion Models

TL;DR

This paper introduces FedDiff, a diffusion model-based approach for one-shot federated learning that addresses data heterogeneity and privacy challenges, improving efficiency and sample quality under differential privacy constraints.

Contribution

It proposes FedDiff, a novel diffusion model method for one-shot federated learning, and introduces Fourier Magnitude Filtering to enhance data quality under differential privacy.

Findings

FedDiff outperforms existing one-shot FL methods in heterogeneous data scenarios.

FMF improves the quality of generated samples under differential privacy.

The approach enhances model performance while maintaining data privacy.

Abstract

Federated learning (FL) enables multiple clients to train models collectively while preserving data privacy. However, FL faces challenges in terms of communication cost and data heterogeneity. One-shot federated learning has emerged as a solution by reducing communication rounds, improving efficiency, and providing better security against eavesdropping attacks. Nevertheless, data heterogeneity remains a significant challenge, impacting performance. This work explores the effectiveness of diffusion models in one-shot FL, demonstrating their applicability in addressing data heterogeneity and improving FL performance. Additionally, we investigate the utility of our diffusion model approach, FedDiff, compared to other one-shot FL methods under differential privacy (DP). Furthermore, to improve generated sample quality under DP settings, we propose a pragmatic Fourier Magnitude Filtering (FMF) method, enhancing the effectiveness of generated data for global model training.