Privacy Amplification Persists under Unlimited Synthetic Data Release

TL;DR

This paper demonstrates that privacy amplification through synthetic data release remains effective even with unlimited data release under certain assumptions, extending previous asymptotic results to more practical settings.

Contribution

It proves that privacy amplification persists with unlimited synthetic data release under bounded-parameter assumptions, improving prior asymptotic bounds.

Findings

Privacy amplification persists with unlimited synthetic data release.

The analysis applies under bounded-parameter assumptions.

Provides structural insights for tighter privacy guarantees.

Abstract

We study privacy amplification by synthetic data release, a phenomenon in which differential privacy guarantees are improved by releasing only synthetic data rather than the private generative model itself. Recent work by Pierquin et al. (2025) established the first formal amplification guarantees for a linear generator, but they apply only in asymptotic regimes where the model dimension far exceeds the number of released synthetic records, limiting their practical relevance. In this work, we show a surprising result: under a bounded-parameter assumption, privacy amplification persists even when releasing an unbounded number of synthetic records, thereby improving upon the bounds of Pierquin et al. (2025). Our analysis provides structural insights that may guide the development of tighter privacy guarantees for more complex release mechanisms.