FedAlign: Federated Domain Generalization with Cross-Client Feature Alignment

2025-01-26

AI Summary

To address weak domain generalization, non-independent and identically distributed (Non-IID) data, and insufficient domain diversity in federated learning, this paper proposes a lightweight privacy-preserving domain generalization framework. Methodologically, it introduces (1) a novel cross-client feature expansion module that enhances feature diversity via selective feature transfer and domain-invariant perturbation; and (2) a dual-stage alignment mechanism—operating jointly on embedding space and prediction outputs—to improve domain invariance without sharing raw data. The framework rigorously satisfies differential privacy guarantees and relies solely on standard federated averaging for communication. Evaluated on multiple cross-domain federated benchmarks, it achieves significant improvements in unseen-domain accuracy while maintaining low communication and computational overhead.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Application Domains: Security
• Computer Vision: Bias, Fairness & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Federated Learning (FL) offers a decentralized paradigm for collaborative model training without direct data sharing, yet it poses unique challenges for Domain Generalization (DG), including strict privacy constraints, non-i.i.d. local data, and limited domain diversity. We introduce FedAlign, a lightweight, privacy-preserving framework designed to enhance DG in federated settings by simultaneously increasing feature diversity and promoting domain invariance. First, a cross-client feature extension module broadens local domain representations through domain-invariant feature perturbation and selective cross-client feature transfer, allowing each client to safely access a richer domain space. Second, a dual-stage alignment module refines global feature learning by aligning both feature embeddings and predictions across clients, thereby distilling robust, domain-invariant features. By integrating these modules, our method achieves superior generalization to unseen domains while maintaining data privacy and operating with minimal computational and communication overhead.

Problem

• Federated Learning
• Domain Generalization
• Privacy Preservation

Innovation

• Federated Domain Generalization
• Cross-client Feature Alignment
• Privacy Preservation