TU Darmstadt / ULB / TUbiblio

ScionFL: Secure Quantized Aggregation for Federated Learning

Ben-Itzhak, Yaniv ; Möllering, Helen ; Pinkas, Benny ; Schneider, Thomas ; Suresh, Ajith ; Tkachenko, Oleksandr ; Vargaftik, Shay ; Weinert, Christian ; Yalame, Hossein ; Yanai, Avishay (2024)
ScionFL: Secure Quantized Aggregation for Federated Learning.
2nd IEEE Conference on Secure and Trustworthy Machine Learning (SaTML'24). Toronto, Canada (09.04.2024 - 11.04.2024)
doi: 10.1109/SaTML59370.2024.00031
Conference or Workshop Item, Bibliographie

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.

Item Type: Conference or Workshop Item
Erschienen: 2024
Creators: Ben-Itzhak, Yaniv ; Möllering, Helen ; Pinkas, Benny ; Schneider, Thomas ; Suresh, Ajith ; Tkachenko, Oleksandr ; Vargaftik, Shay ; Weinert, Christian ; Yalame, Hossein ; Yanai, Avishay
Type of entry: Bibliographie
Title: ScionFL: Secure Quantized Aggregation for Federated Learning
Language: English
Date: 10 May 2024
Publisher: IEEE
Book Title: Proceedings: IEEE Conference on Safe and Trustworthy Machine Learning: SaTML 2024
Collation: 23 Seiten
Event Title: 2nd IEEE Conference on Secure and Trustworthy Machine Learning (SaTML'24)
Event Location: Toronto, Canada
Event Dates: 09.04.2024 - 11.04.2024
DOI: 10.1109/SaTML59370.2024.00031
Corresponding Links:
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.

Additional Information:

\textbfRunner-up distinguished paper award

Divisions: 20 Department of Computer Science
20 Department of Computer Science > Cryptography and Privacy Engineering (ENCRYPTO)
DFG-Collaborative Research Centres (incl. Transregio)
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres
DFG-Graduiertenkollegs
DFG-Graduiertenkollegs > Research Training Group 2050 Privacy and Trust for Mobile Users
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres > CRC 1119: CROSSING – Cryptography-Based Security Solutions: Enabling Trust in New and Next Generation Computing Environments
Date Deposited: 25 Jul 2024 07:40
Last Modified: 25 Jul 2024 07:40
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details