Choice Outweighs Effort: Facilitating Complementary Knowledge Fusion in Federated Learning via Re-calibration and Merit-discrimination

TL;DR

FedMate is a federated learning method that dynamically balances global consistency and local adaptation by re-calibrating knowledge fusion through a bilateral optimization process, improving performance across diverse datasets.

Contribution

The paper introduces FedMate, a novel federated learning approach that employs dynamic global prototypes and merit-based client-side fusion to better handle data heterogeneity.

Findings

FedMate outperforms existing methods on five diverse datasets.

It achieves better balance between generalization and personalization.

Semantic segmentation results confirm real-world scalability.

Abstract

Cross-client data heterogeneity in federated learning induces biases that impede unbiased consensus condensation and the complementary fusion of generalization- and personalization-oriented knowledge. While existing approaches mitigate heterogeneity through model decoupling and representation center loss, they often rely on static and restricted metrics to evaluate local knowledge and adopt global alignment too rigidly, leading to consensus distortion and diminished model adaptability. To address these limitations, we propose FedMate, a method that implements bilateral optimization: On the server side, we construct a dynamic global prototype, with aggregation weights calibrated by holistic integration of sample size, current parameters, and future prediction; a category-wise classifier is then fine-tuned using this prototype to preserve global consistency. On the client side, we introduce complementary classification fusion to enable merit-based discrimination training and incorporate cost-aware feature transmission to balance model performance and communication efficiency. Experiments on five datasets of varying complexity demonstrate that FedMate outperforms state-of-the-art methods in harmonizing generalization and adaptation. Additionally, semantic segmentation experiments on autonomous driving datasets validate the method's real-world scalability.