LSHFed: Robust and Communication-Efficient Federated Learning with Locally-Sensitive Hashing Gradient Mapping

2025-11-03

AI Summary

Federated learning (FL) faces dual threats: inference attacks compromising client privacy and poisoning attacks degrading model robustness; existing defenses often suffer from excessive communication overhead or insufficient detection accuracy. This paper proposes a communication-efficient and robust FL security framework. Its core innovation is a multi-hyperplane locality-sensitive hashing (LSH) scheme that performs irreversible binary encoding of local gradients, enabling lightweight anomaly detection and aggregation verification directly in the hash domain. The method simultaneously preserves gradient privacy and enables reliable identification of malicious updates, substantially reducing communication costs. Experimental results demonstrate that the framework maintains model accuracy even under 50% client collusion attacks. Gradient verification incurs up to three orders of magnitude lower communication overhead compared to baselines, while achieving significantly higher detection accuracy.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated learning (FL) enables collaborative model training across distributed nodes without exposing raw data, but its decentralized nature makes it vulnerable in trust-deficient environments. Inference attacks may recover sensitive information from gradient updates, while poisoning attacks can degrade model performance or induce malicious behaviors. Existing defenses often suffer from high communication and computation costs, or limited detection precision. To address these issues, we propose LSHFed, a robust and communication-efficient FL framework that simultaneously enhances aggregation robustness and privacy preservation. At its core, LSHFed incorporates LSHGM, a novel gradient verification mechanism that projects high-dimensional gradients into compact binary representations via multi-hyperplane locally-sensitive hashing. This enables accurate detection and filtering of malicious gradients using only their irreversible hash forms, thus mitigating privacy leakage risks and substantially reducing transmission overhead. Extensive experiments demonstrate that LSHFed maintains high model performance even when up to 50% of participants are collusive adversaries while achieving up to a 1000x reduction in gradient verification communication compared to full-gradient methods.

Problem

• Enhancing security against inference and poisoning attacks in federated learning
• Reducing communication costs in decentralized model training systems
• Improving malicious gradient detection while preserving participant privacy

Innovation

• Uses locally-sensitive hashing for gradient mapping
• Detects malicious gradients via binary hash representations
• Reduces communication costs while preserving privacy