A naukri.com initiative
Arxiv
1M
96
Image Credit: Arxiv
Bayesian Pseudo Posterior Mechanism for Differentially Private Machine Learning
- Differential privacy (DP) is important for protecting privacy in machine learning applications.
- A new scalable DP mechanism, SWAG-PPM, is proposed for deep learning models.
- SWAG-PPM outperforms the industry standard DP-SGD for a similar privacy budget.
- SWAG-PPM demonstrates its effectiveness on a workplace injury text classification task.
Read Full Article
5 Likes
For uninterrupted reading, download the app