Differentially Private Learning of Geometric Concepts

Haim Kaplan, Yishay Mansour, Yossi Matias, Uri Stemmer

ICML 2019 pp. 3233-3241

/icml/2019/kaplan2019icml-differentially/

Abstract

We present differentially private efficient algorithms for learning union of polygons in the plane (which are not necessarily convex). Our algorithms achieve $(\alpha,\beta)$-PAC learning and $(\epsilon,\delta)$-differential privacy using a sample of size $\tilde{O}\left(\frac{1}{\alpha\epsilon}k\log d\right)$, where the domain is $[d]\times[d]$ and $k$ is the number of edges in the union of polygons.

PDF ICML Semantic Scholar

Cite

Text

Kaplan et al. "Differentially Private Learning of Geometric Concepts." International Conference on Machine Learning, 2019.

Markdown

[Kaplan et al. "Differentially Private Learning of Geometric Concepts." International Conference on Machine Learning, 2019.](https://mlanthology.org/icml/2019/kaplan2019icml-differentially/)

BibTeX

@inproceedings{kaplan2019icml-differentially,
  title     = {{Differentially Private Learning of Geometric Concepts}},
  author    = {Kaplan, Haim and Mansour, Yishay and Matias, Yossi and Stemmer, Uri},
  booktitle = {International Conference on Machine Learning},
  year      = {2019},
  pages     = {3233-3241},
  volume    = {97},
  url       = {https://mlanthology.org/icml/2019/kaplan2019icml-differentially/}
}