LATTEO: A Framework to Support Learning Asynchronously Tempered with Trusted Execution and Obfuscation

2025-02-07

AI Summary

This work addresses critical privacy challenges in asynchronous federated learning (AFL), including sensitive data reconstruction risks induced by gradient leakage, the absence of synchronization coordination, and heightened privacy threats due to high client participation. We propose the first privacy-preserving framework integrating trusted execution environments (TEEs) with lightweight gradient obfuscation. Our approach innovatively designs a decentralized, multi-authority attribute-based encryption scheme for implicit yet verifiable TEE service authentication, eliminating reliance on centralized certificate authorities. Furthermore, we introduce a low-overhead gradient obfuscation mechanism tailored to AFL’s dynamic client participation and straggler tolerance. Experimental results demonstrate that our method reduces structural similarity of reconstructed data by 85% and increases reconstruction error by 400%. Authentication latency is reduced by 1500% compared to RA-TLS, with zero additional communication or computational overhead.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Search and Optimization: Learning to Search

Application Category

• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

The privacy vulnerabilities of the federated learning (FL) paradigm, primarily caused by gradient leakage, have prompted the development of various defensive measures. Nonetheless, these solutions have predominantly been crafted for and assessed in the context of synchronous FL systems, with minimal focus on asynchronous FL. This gap arises in part due to the unique challenges posed by the asynchronous setting, such as the lack of coordinated updates, increased variability in client participation, and the potential for more severe privacy risks. These concerns have stymied the adoption of asynchronous FL. In this work, we first demonstrate the privacy vulnerabilities of asynchronous FL through a novel data reconstruction attack that exploits gradient updates to recover sensitive client data. To address these vulnerabilities, we propose a privacy-preserving framework that combines a gradient obfuscation mechanism with Trusted Execution Environments (TEEs) for secure asynchronous FL aggregation at the network edge. To overcome the limitations of conventional enclave attestation, we introduce a novel data-centric attestation mechanism based on Multi-Authority Attribute-Based Encryption. This mechanism enables clients to implicitly verify TEE-based aggregation services, effectively handle on-demand client participation, and scale seamlessly with an increasing number of asynchronous connections. Our gradient obfuscation mechanism reduces the structural similarity index of data reconstruction by 85% and increases reconstruction error by 400%, while our framework improves attestation efficiency by lowering average latency by up to 1500% compared to RA-TLS, without additional overhead.

Problem

• Address privacy vulnerabilities in asynchronous federated learning.
• Propose a framework combining gradient obfuscation and TEEs.
• Introduce data-centric attestation for secure FL aggregation.

Innovation

• Gradient obfuscation mechanism
• Trusted Execution Environments
• Multi-Authority Attribute-Based Encryption