Approximate Differential Privacy of the $\ell_2$ Mechanism

Abstract

We study the $\ell_2$ mechanism for computing a $d$-dimensional statistic with bounded $\ell_2$ sensitivity under approximate differential privacy. Across a range of privacy parameters, we find that the $\ell_2$ mechanism obtains error approaching that of the Laplace mechanism as $d \to 1$ and approaching that of the Gaussian mechanism as $d \to \infty$; however, it dominates both in between.

Cite

Text

Joseph et al. "Approximate Differential Privacy of the $\ell_2$ Mechanism." Proceedings of the 42nd International Conference on Machine Learning, 2025.

Markdown

[Joseph et al. "Approximate Differential Privacy of the $\ell_2$ Mechanism." Proceedings of the 42nd International Conference on Machine Learning, 2025.](https://mlanthology.org/icml/2025/joseph2025icml-approximate/)

BibTeX

@inproceedings{joseph2025icml-approximate,
  title     = {{Approximate Differential Privacy of the $\ell_2$ Mechanism}},
  author    = {Joseph, Matthew and Kulesza, Alex and Yu, Alexander},
  booktitle = {Proceedings of the 42nd International Conference on Machine Learning},
  year      = {2025},
  pages     = {28377-28392},
  volume    = {267},
  url       = {https://mlanthology.org/icml/2025/joseph2025icml-approximate/}
}