ScionFL: Efficient and Robust Secure Quantized Aggregation

Abstract

Secure aggregation is commonly used in federated learning (FL) to alleviate privacy concerns related to the central aggregator seeing all parameter updates in the clear. Unfortunately, most existing secure aggregation schemes ignore two critical orthogonal research directions that aim to (i) significantly reduce client-server communication and (ii) mitigate the impact of malicious clients. However, both of these additional properties are essential to facilitate cross-device FL with thousands or even millions of (mobile) participants. In this paper, we unite both research directions by introducing ScionFL, the first secure aggregation framework for FL that operates efficiently on quantized inputs and simultaneously provides robustness against malicious clients. Our framework leverages (novel) multi-party computation (MPC) techniques and supports multiple linear (1-bit) quantization schemes, including ones that utilize the randomized Hadamard transform and Kashin's representation. Our theoretical results are supported by extensive evaluations. We show that with no overhead for clients and moderate overhead for the server compared to transferring and processing quantized updates in plaintext, we obtain comparable accuracy for standard FL benchmarks. Moreover, we demonstrate the robustness of our framework against state-of-the-art poisoning attacks.

Published in 2024 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)

References in corpus (19)

• Federated Learning: Strategies for Improving Communication Efficiency
• Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
• Sparse Communication for Distributed Gradient Descent
• Can You Really Backdoor Federated Learning?
• A Field Guide to Federated Optimization
• XONN: XNOR-based Oblivious Deep Neural Network Inference
• Trident: Efficient 4PC Framework for Privacy Preserving Machine Learning
• FastSecAgg: Scalable Secure Aggregation for Privacy-Preserving Federated Learning
• ASTRA: High Throughput 3PC over Rings with Application to Secure Prediction
• Uncertainty Principles and Vector Quantization
• Tetrad: Actively Secure 4PC for Secure Training and Inference
• EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback
• SWIFT: Super-fast and Robust Privacy-Preserving Machine Learning
• Eavesdrop the Composition Proportion of Training Labels in Federated Learning
• When the Curious Abandon Honesty: Federated Learning Is Not Private
• BLAZE: Blazing Fast Privacy-Preserving Machine Learning
• Efficient Sparse Secure Aggregation for Federated Learning
• SCOTCH: An Efficient Secure Computation Framework for Secure Aggregation
• QUIC-FL: Quick Unbiased Compression for Federated Learning