On the Formal Privacy Guarantees of Synthetic Data (Generated Without Formal Privacy Guarantees) (im Erscheinen)

Abstract

"What privacy guarantees can synthetic data satisfy, even without formal guarantees, during the training of the synthesizer? In this paper, we explore this question using a synthesizer under simplified settings to show that the privacy guarantees offered by this synthesizer and potential others can be directly translated into a \rho-zCDP guarantee. We further explore the conditions under which this equivalence holds and show that getting formal privacy guarantees for more realistic synthetic data models is significantly harder." (Author's abstract, IAB-Doku) ((en))

Cite article

Neunhoeffer, M., Seeman, J. & Drechsler, J. (2025): On the Formal Privacy Guarantees of Synthetic Data (Generated Without Formal Privacy Guarantees) (im Erscheinen). In: Harvard Data Science Review.