Robust Blockchained Federated Learning with Model Validation and Proof-of-Stake Inspired Consensus

TL;DR

This paper presents VBFL, a blockchain-based federated learning framework that enhances robustness against malicious devices through decentralized validation and a proof-of-stake consensus, significantly improving accuracy in adversarial settings.

Contribution

Introduces a novel blockchain-based decentralized FL framework with validation and proof-of-stake mechanisms to improve robustness and security.

Findings

VBFL achieves 87% accuracy with 15% malicious devices.

Outperforms vanilla FL by 7.4 times in accuracy under attack.

Demonstrates robustness of the proposed framework on MNIST dataset.

Abstract

Federated learning (FL) is a promising distributed learning solution that only exchanges model parameters without revealing raw data. However, the centralized architecture of FL is vulnerable to the single point of failure. In addition, FL does not examine the legitimacy of local models, so even a small fraction of malicious devices can disrupt global training. To resolve these robustness issues of FL, in this paper, we propose a blockchain-based decentralized FL framework, termed VBFL, by exploiting two mechanisms in a blockchained architecture. First, we introduced a novel decentralized validation mechanism such that the legitimacy of local model updates is examined by individual validators. Second, we designed a dedicated proof-of-stake consensus mechanism where stake is more frequently rewarded to honest devices, which protects the legitimate local model updates by increasing their chances of dictating the blocks appended to the blockchain. Together, these solutions promote more federation within legitimate devices, enabling robust FL. Our emulation results of the MNIST classification corroborate that with 15% of malicious devices, VBFL achieves 87% accuracy, which is 7.4x higher than Vanilla FL.