Federated Synthetic Data Generation with Stronger Security Guarantees

Federated Synthetic Data Generation with Stronger Security Guarantees

Ghavamipour, Ali Reza (University Medical Center Groningen)
Turkmen, Fatih (University Medical Center Groningen)
Wang, Rui (Student TU Delft)
Liang, K. (TU Delft Cyber Security)

2023

Synthetic data generation plays a crucial role in many areas where data is scarce and privacy/confidentiality is a significant concern. Generative Adversarial Networks (GANs), arguably one of the most widely used data synthesis techniques, allow for the training of a model (i.e., generator) that can generate real-looking data by playing a min-max game with a discriminator model. When multiple organizations are reluctant to share their sensitive data, GANs models can be trained in a federated manner, commonly with the use of differential privacy (DP). In order to achieve a reasonable level of model utility, DP trades privacy exhibiting vulnerability to various attacks (e.g., membership inference attack). In this paper, we propose a hybrid solution, PP-FedGAN, to the asynchronous federated, privacy-preserving training of GANs models by combining the CKKS homomorphic encryption (HE) scheme with differential privacy. The addition of HE results in around 10 seconds of overhead on the client side per round and 115 seconds on the entire training procedure. We also analyze the security of PP-FedGAN under the honest-but-curious security model. Where stronger security guarantees are required, our proposal presents a better alternative to solutions that only employ DP.

differential privacy
federated learning
gan
homomorphic encryption
synthetic data

http://resolver.tudelft.nl/uuid:2cba0f7e-b1ae-4e69-a0f1-153df866891c

https://doi.org/10.1145/3589608.3593835

Association for Computing Machinery (ACM)

979-8-4007-0173-3

SACMAT 2023 - Proceedings of the 28th ACM Symposium on Access Control Models and Technologies

28th ACM Symposium on Access Control Models and Technologies, SACMAT 2023, 2023-06-07 → 2023-06-09, Trento, Italy

Proceedings of ACM Symposium on Access Control Models and Technologies, SACMAT

Institutional Repository

conference paper

© 2023 Ali Reza Ghavamipour, Fatih Turkmen, Rui Wang, K. Liang