Practical Differentially Private and Byzantine-resilient Federated Learning
Open Access
- 13 June 2023
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 1 (2) , 1-26
Abstract
Privacy and Byzantine resilience are two indispensable requirements for a federated learning (FL) system. Although there have been extensive studies on privacy and Byzantine security in their own track, solutions that consider both remain sparse. This is due to difficulties in reconciling privacy-preserving and Byzantine-resilient algorithms. In this work, we propose a solution to such a two-fold issue. We use our version of differentially private stochastic gradient descent (DP-SGD) algorithm to preserve privacy and then apply our Byzantine-resilient algorithms. We note that while existing works follow this general approach, an in-depth analysis on the interplay between DP and Byzantine resilience has been ignored, leading to unsatisfactory performance. Specifically, for the random noise introduced by DP, previous works strive to reduce its seemingly detrimental impact on the Byzantine aggregation. In contrast, we leverage the random noise to construct a first-stage aggregation that effectively rejects many existing Byzantine attacks. Moreover, based on another property of our DP variant, we form a second-stage aggregation which provides a final sound filtering. Our protocol follows the principle of co-designing both DP and Byzantine resilience. We provide both theoretical proof and empirical experiments to show our protocol is effective: retaining high accuracy while preserving the DP guarantee and Byzantine resilience. Compared with the previous work, our protocol 1) achieves significantly higher accuracy even in a high privacy regime; 2) works well even when up to 90% distributive workers are Byzantine.Keywords
All Related Versions
Funding Information
- National Science Foundation (2220433 2213700)
- KAUST-SDAIA Center of Excellence in Data Science and Artificial Intelligence (REI/1/4811-10-01, BAS/1/1689-01-01 URF/1/4663-01-01 FCC/1/1976-49-01 RGC/3/4816)
This publication has 23 references indexed in Scilit:
- Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning2018
- Distributed Statistical Machine Learning in Adversarial SettingsProceedings of the ACM on Measurement and Analysis of Computing Systems, 2017
- Membership Inference Attacks Against Machine Learning Models2017
- Is Interaction Necessary for Distributed Private Learning?2017
- Deep Learning with Differential Privacy2016
- Stochastic gradient descent with differentially private updates2013
- ANTIDOTE2009
- Evaluating Kolmogorov's DistributionJournal of Statistical Software, 2003
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- A database for handwritten text recognition researchIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994