A Secure Federated Learning: Analysis of Different Cryptographic Tools

Oana Stan, Vincent Thouvenot, Aymen Boudguiga, Katarzyna Kapusta, Martin Zuber, Renaud Sirdey

2022

Abstract

Federated Learning is established as one of the most efficient collaborative learning approaches aiming at training different client models using private datasets. By private, we mean that clients’ datasets are never disclosed as they serve to train clients’ models locally. Then, a central server is in charge of aggregating the different models’ weights. The central server is generally a honest-but-curious entity that may be interested in collecting information about clients datasets by using model inversion or membership inference. In this paper, we discuss different cryptographic options for providing a secure Federated Learning framework. We investigate the use of Differential Privacy, Homomorphic Encryption and Multi-Party Computation (MPC) for confidential data aggregation while considering different threat models. In our homomorphic encryption approach, we compare results obtained with an optimized version of the Paillier cryptosystem to those obtained with BFV and CKKS. As for MPC technique, different general protocols are tested under various security assumptions. Overall we have found HE to have better performance, for a lower bandwidth usage.

Download


Paper Citation


in Harvard Style

Stan O., Thouvenot V., Boudguiga A., Kapusta K., Zuber M. and Sirdey R. (2022). A Secure Federated Learning: Analysis of Different Cryptographic Tools. In Proceedings of the 19th International Conference on Security and Cryptography - Volume 1: SECRYPT, ISBN 978-989-758-590-6, pages 669-674. DOI: 10.5220/0011322700003283


in Bibtex Style

@conference{secrypt22,
author={Oana Stan and Vincent Thouvenot and Aymen Boudguiga and Katarzyna Kapusta and Martin Zuber and Renaud Sirdey},
title={A Secure Federated Learning: Analysis of Different Cryptographic Tools},
booktitle={Proceedings of the 19th International Conference on Security and Cryptography - Volume 1: SECRYPT,},
year={2022},
pages={669-674},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011322700003283},
isbn={978-989-758-590-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Conference on Security and Cryptography - Volume 1: SECRYPT,
TI - A Secure Federated Learning: Analysis of Different Cryptographic Tools
SN - 978-989-758-590-6
AU - Stan O.
AU - Thouvenot V.
AU - Boudguiga A.
AU - Kapusta K.
AU - Zuber M.
AU - Sirdey R.
PY - 2022
SP - 669
EP - 674
DO - 10.5220/0011322700003283