Lower Bounds for Private Estimation of Gaussian Covariance Matrices under All Reasonable Parameter Regimes
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929329841111040 |
|---|---|
| author | Portella, Victor S. Harvey, Nick |
| author_facet | Portella, Victor S. Harvey, Nick |
| contents | We prove lower bounds on the number of samples needed to privately estimate the covariance matrix of a Gaussian distribution. Our bounds match existing upper bounds in the widest known setting of parameters. Our analysis relies on the Stein-Haff identity, an extension of the classical Stein's identity used in previous fingerprinting lemma arguments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_17714 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Lower Bounds for Private Estimation of Gaussian Covariance Matrices under All Reasonable Parameter Regimes Portella, Victor S. Harvey, Nick Data Structures and Algorithms Cryptography and Security Machine Learning We prove lower bounds on the number of samples needed to privately estimate the covariance matrix of a Gaussian distribution. Our bounds match existing upper bounds in the widest known setting of parameters. Our analysis relies on the Stein-Haff identity, an extension of the classical Stein's identity used in previous fingerprinting lemma arguments. |
| title | Lower Bounds for Private Estimation of Gaussian Covariance Matrices under All Reasonable Parameter Regimes |
| topic | Data Structures and Algorithms Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2404.17714 |