BRFL: A blockchain-based byzantine-robust federated learning model

• Published in: Journal of parallel and distributed computing Vol. 196; p. 104995
• Main Authors: Li, Yang, Xia, Chunhe, Li, Chang, Wang, Tianbo
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.02.2025
• Subjects: Blockchain; Federated learning; Pearson correlation coefficient; Spectral clustering; Spectral clustering; Federated learning; Blockchain; Pearson correlation coefficient
• ISSN: 0743-7315
• DOI: 10.1016/j.jpdc.2024.104995

With the increasing importance of machine learning, the privacy and security of training data have become a concern. Federated learning, which stores data in distributed nodes and shares only model parameters, has gained significant attention for addressing this concern. However, a challenge arises in federated learning due to the byzantine attack problem, where malicious local models can compromise the global model's performance during aggregation. This article proposes the Blockchain-based Byzantine-Robust Federated Learning (BRFL) model, which combines federated learning with blockchain technology. We improve the robustness of federated learning by proposing a new consensus algorithm and aggregation algorithm for blockchain-based federated learning. Meanwhile, we modify the block saving rules of the blockchain to reduce the storage pressure of the nodes. Experimental results on public datasets demonstrate the superior byzantine robustness of our secure aggregation algorithm compared to other baseline aggregation methods, and reduce the storage pressure of the blockchain nodes. •We construct a blockchain-based federated learning model, select trustworthy aggregation nodes in each training round.•Cluster based on the linear similarity between local models, and verify the accuracy of each cluster's local model.•The identification of the transaction type contained in the block is added to the block header.•Experiments demonstrate that our aggregation method outperforms baseline approaches in terms of accuracy.