FedDBP: Enhancing Federated Prototype Learning with Dual-Branch Features and Personalized Global Fusion

TL;DR

FedDBP introduces a dual-branch feature projector and personalized global prototype fusion to improve federated prototype learning, effectively balancing feature fidelity and discriminability amid heterogeneity.

Contribution

The paper proposes FedDBP, a novel federated prototype learning method with dual-branch feature projection and Fisher information-based prototype fusion, addressing existing limitations.

Findings

FedDBP outperforms ten existing methods in experiments.

Dual-branch projector ensures feature fidelity and discriminability.

Fisher information-based fusion effectively identifies important prototype channels.

Abstract

Federated prototype learning (FPL), as a solution to heterogeneous federated learning (HFL), effectively alleviates the challenges of data and model heterogeneity.However, existing FPL methods fail to balance the fidelity and discriminability of the feature, and are limited by a single global prototype. In this paper, we propose FedDBP, a novel FPL method to address the above issues. On the client-side, we design a Dual-Branch feature projector that employs L2 alignment and contrastive learning simultaneously, thereby ensuring both the fidelity and discriminability of local features. On the server-side, we introduce a Personalized global prototype fusion approach that leverages Fisher information to identify the important channels of local prototypes. Extensive experiments demonstrate the superiority of FedDBP over ten existing advanced methods.