TAPAS: A Toolbox for Adversarial Privacy Auditing of Synthetic Data

Abstract

Personal data collected at scale promises to improve decision-making and accelerate innovation. However, sharing and using such data raises serious privacy concerns. A promising solution is to produce synthetic data, artificial records to share instead of real data. Since synthetic records are not linked to real persons, this intuitively prevents classical re-identification attacks. However, this is insufficient to protect privacy. We here present PrivE, a toolbox of attacks to evaluate synthetic data privacy under a wide range of scenarios. These attacks include generalizations of prior works and novel attacks. We also introduce a general framework for reasoning about privacy threats to synthetic data and showcase PrivE on several examples.

Cite

Text

Houssiau et al. "TAPAS: A Toolbox for Adversarial Privacy Auditing of Synthetic Data." NeurIPS 2022 Workshops: SyntheticData4ML, 2022.

Markdown

[Houssiau et al. "TAPAS: A Toolbox for Adversarial Privacy Auditing of Synthetic Data." NeurIPS 2022 Workshops: SyntheticData4ML, 2022.](https://mlanthology.org/neuripsw/2022/houssiau2022neuripsw-tapas/)

BibTeX

@inproceedings{houssiau2022neuripsw-tapas,
  title     = {{TAPAS: A Toolbox for Adversarial Privacy Auditing of Synthetic Data}},
  author    = {Houssiau, Florimond and Jordon, James and Cohen, Samuel N and Daniel, Owen and Elliott, Andrew and Geddes, James and Mole, Callum and Rangel-Smith, Camila and Szpruch, Lukasz},
  booktitle = {NeurIPS 2022 Workshops: SyntheticData4ML},
  year      = {2022},
  url       = {https://mlanthology.org/neuripsw/2022/houssiau2022neuripsw-tapas/}
}