Coupling public and private gradient provably helps optimization

[PDF] arxiv.org

Pre-training Differentially Private Models with Limited Public Data

Z Bu, X Zhang, M Hong, S Zha, G Karypis - arXiv preprint arXiv …, 2024 - arxiv.org
The superior performance of large foundation models relies on the use of massive amounts
of high-quality data, which often contain sensitive, private and copyrighted material that …
被引用次数：2 相关文章 所有 2 个版本
[PDF] arxiv.org

Locally Private Sampling with Public Data

B Zamanlooy, M Diaz, S Asoodeh - arXiv preprint arXiv:2411.08791, 2024 - arxiv.org
Local differential privacy (LDP) is increasingly employed in privacy-preserving machine
learning to protect user data before sharing it with an untrusted aggregator. Most LDP …
相关文章 所有 2 个版本