Privacy Amplification Persists under Unlimited Synthetic Data Release
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915775128797184 |
|---|---|
| author | Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu |
| author_facet | Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_04895 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Privacy Amplification Persists under Unlimited Synthetic Data Release Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu Cryptography and Security Data Structures and Algorithms Machine Learning 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. |
| title | Privacy Amplification Persists under Unlimited Synthetic Data Release |
| topic | Cryptography and Security Data Structures and Algorithms Machine Learning |
| url | https://arxiv.org/abs/2602.04895 |