Cross-Silo Prototypical Calibration for Federated Learning with Non-IID Data

TL;DR

This paper introduces FedCSPC, a novel federated learning calibration method that uses prototype information and contrastive learning to address data heterogeneity and improve model performance across clients.

Contribution

The paper proposes a cross-silo prototypical calibration approach with data prototypical modeling and contrastive learning, enhancing feature consistency in non-IID federated data.

Findings

FedCSPC outperforms state-of-the-art methods on multiple datasets.

The method effectively aligns features across different data sources.

Calibration improves model robustness and generalization.

Abstract

Federated Learning aims to learn a global model on the server side that generalizes to all clients in a privacy-preserving manner, by leveraging the local models from different clients. Existing solutions focus on either regularizing the objective functions among clients or improving the aggregation mechanism for the improved model generalization capability. However, their performance is typically limited by the dataset biases, such as the heterogeneous data distributions and the missing classes. To address this issue, this paper presents a cross-silo prototypical calibration method (FedCSPC), which takes additional prototype information from the clients to learn a unified feature space on the server side. Specifically, FedCSPC first employs the Data Prototypical Modeling (DPM) module to learn data patterns via clustering to aid calibration. Subsequently, the cross-silo prototypical calibration (CSPC) module develops an augmented contrastive learning method to improve the robustness of the calibration, which can effectively project cross-source features into a consistent space while maintaining clear decision boundaries. Moreover, the CSPC module's ease of implementation and plug-and-play characteristics make it even more remarkable. Experiments were conducted on four datasets in terms of performance comparison, ablation study, in-depth analysis and case study, and the results verified that FedCSPC is capable of learning the consistent features across different data sources of the same class under the guidance of calibrated model, which leads to better performance than the state-of-the-art methods. The source codes have been released at https://github.com/qizhuang-qz/FedCSPC.