Privacy Amplification Through Synthetic Data: Insights from Linear Regression
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909639498530816 |
|---|---|
| author | Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu |
| author_facet | Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu |
| contents | Synthetic data inherits the differential privacy guarantees of the model used to generate it. Additionally, synthetic data may benefit from privacy amplification when the generative model is kept hidden. While empirical studies suggest this phenomenon, a rigorous theoretical understanding is still lacking. In this paper, we investigate this question through the well-understood framework of linear regression. First, we establish negative results showing that if an adversary controls the seed of the generative model, a single synthetic data point can leak as much information as releasing the model itself. Conversely, we show that when synthetic data is generated from random inputs, releasing a limited number of synthetic data points amplifies privacy beyond the model's inherent guarantees. We believe our findings in linear regression can serve as a foundation for deriving more general bounds in the future. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05101 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Privacy Amplification Through Synthetic Data: Insights from Linear Regression Pierquin, Clément Bellet, Aurélien Tommasi, Marc Boussard, Matthieu Machine Learning Cryptography and Security Synthetic data inherits the differential privacy guarantees of the model used to generate it. Additionally, synthetic data may benefit from privacy amplification when the generative model is kept hidden. While empirical studies suggest this phenomenon, a rigorous theoretical understanding is still lacking. In this paper, we investigate this question through the well-understood framework of linear regression. First, we establish negative results showing that if an adversary controls the seed of the generative model, a single synthetic data point can leak as much information as releasing the model itself. Conversely, we show that when synthetic data is generated from random inputs, releasing a limited number of synthetic data points amplifies privacy beyond the model's inherent guarantees. We believe our findings in linear regression can serve as a foundation for deriving more general bounds in the future. |
| title | Privacy Amplification Through Synthetic Data: Insights from Linear Regression |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2506.05101 |