• español
    • English
  • Login
  • English 
    • español
    • English
  • Publication Types
    • bookbook partconference objectdoctoral thesisjournal articlemagazinemaster thesispatenttechnical documentationtechnical report
View Item 
  •   IMDEA Networks Home
  • View Item
  •   IMDEA Networks Home
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

MUDGUARD: Taming Malicious Majorities in Federated Learning using Privacy-Preserving Byzantine-Robust Clustering

Share
Files
2208.10161v2.pdf (15.67Mb)
Identifiers
URI: https://hdl.handle.net/20.500.12761/1865
Metadata
Show full item record
Author(s)
Wang, Rui; Wang, Xingkai; Chen, Huanhuan; Decouchant, Jérémie; Picek, Stjepan; Laoutaris, Nikolaos; Liang, Kaitai
Date
2025-06-09
Abstract
Byzantine-robust Federated Learning (FL) aims to counter malicious clients and train an accurate global model while maintaining an extremely low attack success rate. Most existing systems, however, are only robust when most of the clients are honest. \texttt{FLTrust} (NDSS '21) and \texttt{Zeno++} (ICML '20) do not make such an honest majority assumption but can only be applied to scenarios where the server is provided with an auxiliary dataset used to filter malicious updates. \texttt{FLAME} (USENIX '22) and \texttt{EIFFeL} (CCS '22) maintain the semi-honest majority assumption to guarantee robustness and the confidentiality of updates. It is, therefore, currently impossible to ensure Byzantine robustness and confidentiality of updates without assuming a semi-honest majority. To tackle this problem, we propose a novel Byzantine-robust and privacy-preserving FL system, called \texttt{MUDGUARD}, to capture malicious minority and majority for server and client sides, respectively. Our experimental results demonstrate that the accuracy of \texttt{MUDGUARD} is practically close to the FL baseline using FedAvg without attacks ($\approx$0.8\% gap on average). Meanwhile, the attack success rate is around 0\%-5\% even under an adaptive attack tailored to \texttt{MUDGUARD}. We further optimize our design by using binary secret sharing and polynomial transformation, leading to communication overhead and runtime decreases of 67\%-89.17\% and 66.05\%-68.75\%, respectively.
Share
Files
2208.10161v2.pdf (15.67Mb)
Identifiers
URI: https://hdl.handle.net/20.500.12761/1865
Metadata
Show full item record

Browse

All of IMDEA NetworksBy Issue DateAuthorsTitlesKeywordsTypes of content

My Account

Login

Statistics

View Usage Statistics

Dissemination

emailContact person Directory wifi Eduroam rss_feed News
IMDEA initiative About IMDEA Networks Organizational structure Annual reports Transparency
Follow us in:
Community of Madrid

EUROPEAN UNION

European Social Fund

EUROPEAN UNION

European Regional Development Fund

EUROPEAN UNION

European Structural and Investment Fund

© 2021 IMDEA Networks. | Accesibility declaration | Privacy Policy | Disclaimer | Cookie policy - We value your privacy: this site uses no cookies!