Federated Synthetic Data Generation with Stronger Security Guarantees

Ali Reza Ghavamipour, Fatih Turkmen, Rui Wang, Kaitai Liang

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

1 Citation (Scopus)
78 Downloads (Pure)

Abstract

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.

Original languageEnglish
Title of host publicationSACMAT 2023 - Proceedings of the 28th ACM Symposium on Access Control Models and Technologies
PublisherAssociation for Computing Machinery (ACM)
Pages31-42
Number of pages12
ISBN (Electronic)979-8-4007-0173-3
DOIs
Publication statusPublished - 2023
Event28th ACM Symposium on Access Control Models and Technologies, SACMAT 2023 - Trento, Italy
Duration: 7 Jun 20239 Jun 2023

Publication series

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

Conference

Conference28th ACM Symposium on Access Control Models and Technologies, SACMAT 2023
Country/TerritoryItaly
CityTrento
Period7/06/239/06/23

Keywords

  • differential privacy
  • federated learning
  • gan
  • homomorphic encryption
  • synthetic data

Fingerprint

Dive into the research topics of 'Federated Synthetic Data Generation with Stronger Security Guarantees'. Together they form a unique fingerprint.

Cite this