Hi-SAFE: Hierarchical Secure Aggregation for Lightweight Federated Learning

2025-11-24

AI Summary

Federated learning in IoT and edge networks faces dual challenges of privacy leakage (e.g., sign-based gradient inference attacks) and excessive communication overhead. Method: This paper proposes the first lightweight, cryptographically secure aggregation framework compatible with sign-based gradient transmission (e.g., SIGNSGD). It constructs a constant-depth majority-voting polynomial via Fermat’s Little Theorem, integrated with hierarchical subgroup partitioning and low-degree polynomial encoding over finite fields to securely aggregate gradient signs. Crucially, it requires no modification to clients’ local update procedures. Contribution/Results: The framework enables provably secure multi-party computation with communication and computational costs independent of client scale. Experiments demonstrate sublinear complexity even in million-device settings—significantly outperforming existing secure aggregation schemes—and establish a new paradigm for jointly optimizing privacy and efficiency in resource-constrained environments.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Multiagent Systems: Mechanism Design

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Federated learning (FL) faces challenges in ensuring both privacy and communication efficiency, particularly in resource-constrained environments such as Internet of Things (IoT) and edge networks. While sign-based methods, such as sign stochastic gradient descent with majority voting (SIGNSGD-MV), offer substantial bandwidth savings, they remain vulnerable to inference attacks due to exposure of gradient signs. Existing secure aggregation techniques are either incompatible with sign-based methods or incur prohibitive overhead. To address these limitations, we propose Hi-SAFE, a lightweight and cryptographically secure aggregation framework for sign-based FL. Our core contribution is the construction of efficient majority vote polynomials for SIGNSGD-MV, derived from Fermat's Little Theorem. This formulation represents the majority vote as a low-degree polynomial over a finite field, enabling secure evaluation that hides intermediate values and reveals only the final result. We further introduce a hierarchical subgrouping strategy that ensures constant multiplicative depth and bounded per-user complexity, independent of the number of users n.

Problem

• Ensuring privacy and communication efficiency in federated learning
• Protecting sign-based methods from inference attacks on gradient signs
• Reducing overhead of secure aggregation for resource-constrained environments

Innovation

• Secure majority vote polynomials from Fermat's Little Theorem
• Hierarchical subgrouping for constant multiplicative depth
• Lightweight cryptographic aggregation for sign-based federated learning