Training generative models from privatized data
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909124273373184 |
|---|---|
| author | Reshetova, Daria Chen, Wei-Ning Özgür, Ayfer |
| author_facet | Reshetova, Daria Chen, Wei-Ning Özgür, Ayfer |
| contents | Local differential privacy is a powerful method for privacy-preserving data collection. In this paper, we develop a framework for training Generative Adversarial Networks (GANs) on differentially privatized data. We show that entropic regularization of optimal transport - a popular regularization method in the literature that has often been leveraged for its computational benefits - enables the generator to learn the raw (unprivatized) data distribution even though it only has access to privatized samples. We prove that at the same time this leads to fast statistical convergence at the parametric rate. This shows that entropic regularization of optimal transport uniquely enables the mitigation of both the effects of privatization noise and the curse of dimensionality in statistical convergence. We provide experimental evidence to support the efficacy of our framework in practice. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_09547 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Training generative models from privatized data Reshetova, Daria Chen, Wei-Ning Özgür, Ayfer Machine Learning Cryptography and Security Information Theory Local differential privacy is a powerful method for privacy-preserving data collection. In this paper, we develop a framework for training Generative Adversarial Networks (GANs) on differentially privatized data. We show that entropic regularization of optimal transport - a popular regularization method in the literature that has often been leveraged for its computational benefits - enables the generator to learn the raw (unprivatized) data distribution even though it only has access to privatized samples. We prove that at the same time this leads to fast statistical convergence at the parametric rate. This shows that entropic regularization of optimal transport uniquely enables the mitigation of both the effects of privatization noise and the curse of dimensionality in statistical convergence. We provide experimental evidence to support the efficacy of our framework in practice. |
| title | Training generative models from privatized data |
| topic | Machine Learning Cryptography and Security Information Theory |
| url | https://arxiv.org/abs/2306.09547 |