FedSCAl: Leveraging Server and Client Alignment for Unsupervised Federated Source-Free Domain Adaptation

TL;DR

This paper introduces FedSCAl, a federated learning framework that uses server-client prediction alignment to improve unsupervised domain adaptation across clients with significant domain gaps, without access to source data.

Contribution

FedSCAl is the first framework to leverage server-client alignment for source-free federated domain adaptation, reducing client drift and improving pseudo-label accuracy.

Findings

FedSCAl outperforms existing FL methods on benchmark datasets.

Server-client alignment improves pseudo-label accuracy.

The method effectively handles data heterogeneity in FFreeDA.

Abstract

We address the Federated source-Free Domain Adaptation (FFreeDA) problem, with clients holding unlabeled data with significant inter-client domain gaps. The FFreeDA setup constrains the FL frameworks to employ only a pre-trained server model as the setup restricts access to the source dataset during the training rounds. Often, this source domain dataset has a distinct distribution to the clients' domains. To address the challenges posed by the FFreeDA setup, adaptation of the Source-Free Domain Adaptation (SFDA) methods to FL struggles with client-drift in real-world scenarios due to extreme data heterogeneity caused by the aforementioned domain gaps, resulting in unreliable pseudo-labels. In this paper, we introduce FedSCAl, an FL framework leveraging our proposed Server-Client Alignment (SCAl) mechanism to regularize client updates by aligning the clients' and server model's predictions. We observe an improvement in the clients' pseudo-labeling accuracy post alignment, as the SCAl mechanism helps to mitigate the client-drift. Further, we present extensive experiments on benchmark vision datasets showcasing how FedSCAl consistently outperforms state-of-the-art FL methods in the FFreeDA setup for classification tasks.