FedGCA: Global Consistent Augmentation Based Single-Source Federated Domain Generalization

TL;DR

FedGCA introduces a style augmentation approach with consistency constraints to improve federated domain generalization, enabling models to better generalize to unseen domains despite single-source, non-IID data limitations.

Contribution

The paper proposes FedGCA, a novel style augmentation method with consistency constraints, to enhance cross-domain generalization in federated learning from single-source data.

Findings

FedGCA outperforms existing methods in federated domain generalization tasks.

The style-complement module effectively diversifies domain styles.

Consistency constraints improve the integration of augmented samples.

Abstract

Federated Domain Generalization (FedDG) aims to train the global model for generalization ability to unseen domains with multi-domain training samples. However, clients in federated learning networks are often confined to a single, non-IID domain due to inherent sampling and temporal limitations. The lack of cross-domain interaction and the in-domain divergence impede the learning of domain-common features and limit the effectiveness of existing FedDG, referred to as the single-source FedDG (sFedDG) problem. To address this, we introduce the Federated Global Consistent Augmentation (FedGCA) method, which incorporates a style-complement module to augment data samples with diverse domain styles. To ensure the effective integration of augmented samples, FedGCA employs both global guided semantic consistency and class consistency, mitigating inconsistencies from local semantics within individual clients and classes across multiple clients. The conducted extensive experiments demonstrate the superiority of FedGCA.