FedAlign: Federated Domain Generalization with Cross-Client Feature Alignment
This addresses domain generalization challenges for federated learning applications, such as healthcare or IoT, where data privacy and diverse client data are critical, representing an incremental improvement over existing methods.
The paper tackled the problem of domain generalization in federated learning, where privacy constraints and non-i.i.d. data limit domain diversity, by introducing FedAlign, a framework that increased feature diversity and promoted domain invariance, achieving superior generalization to unseen domains with minimal overhead.
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.