JAX-Privacy: A library for differentially private machine learning
Not on the map. Why?
Abstract (arXiv)
JAX-Privacy is a library designed to simplify the deployment of robust and performant mechanisms for differentially private machine learning. Guided by design principles of usability, flexibility, and efficiency, JAX-Privacy serves both researchers requiring deep customization and practitioners who want a more out-of-the-box experience. The library provides verified, modular primitives for critical components for all aspects of the mechanism design including batch selection, gradient clipping, noise addition, accounting, and auditing, and brings together a large body of recent research on differentially private ML.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 19 of its 25 references that are in the map. First posted on arXiv 19 Feb 2026.
Builds on
19 of 25 references.
References · 19
- 2016Deep Learning with Differential Privacycited by 2,682
- 2021Opacus: User-Friendly Differential Privacy Library in PyTorchcited by 375
- 2018Privacy Amplification by Subsampling: Tight Analyses via Couplings and Divergencescited by 237
- 2020Auditing Differentially Private Machine Learning: How Private is Private SGD?cited by 213
- 2021Adversary Instantiation: Lower Bounds for Differentially Private Machine Learningcited by 193
- 2023Privacy Auditing with One (1) Training Runcited by 119
- 2022Differentially Private Optimization on Large Model at Small Costcited by 65
- 2023(Amplified) Banded Matrix Factorization: A unified approach to private trainingcited by 49
- 2020Scaling up Differentially Private Deep Learning with Fast Per-Example Gradient Clippingcited by 46
- 2022Debugging Differential Privacy: A Case Study for Privacy Auditingcited by 39
- 2024Fine-Tuning Large Language Models with User-Level Differential Privacycited by 28
- 2024Auditing f-Differential Privacy in One Runcited by 28
7 more
- 2024Near Exact Privacy Amplification for Matrix Mechanismscited by 17
- 2024Scalable DP-SGD: Shuffling vs. Poisson Subsamplingcited by 16
- 2025VaultGemma: A Differentially Private Gemma Modelcited by 14
- 2024Balls-and-Bins Sampling for DP-SGDcited by 13
- 2025Correlated Noise Mechanisms for Differentially Private Learningcited by 13
- 2023Sparsity-Preserving Differentially Private Training of Large Embedding Modelscited by 5
- 2025Tighter Privacy Analysis for Truncated Poisson Samplingcited by 5
Built on by
Citing papers per year: 2026: 2.
2 papers.