UAI 2020poster13 citations

Differentially Private Small Dataset Release Using Random Projections

Lovedeep Gondara, Ke Wang

Abstract

Small datasets form a significant portion of releasable data in high sensitivity domains such as healthcare. But, providing differential privacy for small dataset release is a hard task, where current state-of-the-art methods suffer from severe utility loss. As a solution, we propose DPRP (Differentially Private Data Release via Random Projections), a reconstruction based approach for releasing differentially private small datasets. DPRP has several key advantages over the state-of-the-art. Using seven diverse real-life datasets, we show that DPRP outperforms the current state-of-the-art on a variety of tasks, under varying conditions, and for all privacy budgets.

BibTeX
@InProceedings{pmlr-v124-gondara20a,
  title = 	 {Differentially Private Small Dataset Release Using Random Projections},
  author =       {Gondara, Lovedeep and Wang, Ke},
  booktitle = 	 {Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI)},
  pages = 	 {639--648},
  year = 	 {2020},
  editor = 	 {Peters, Jonas and Sontag, David},
  volume = 	 {124},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {03--06 Aug},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v124/gondara20a/gondara20a.pdf},
  url = 	 {https://proceedings.mlr.press/v124/gondara20a.html},
  abstract = 	 {Small datasets form a significant portion of releasable data in high sensitivity domains such as healthcare. But, providing differential privacy for small dataset release is a hard task, where current state-of-the-art methods suffer from severe utility loss. As a solution, we propose DPRP (Differentially Private Data Release via Random Projections), a reconstruction based approach for releasing differentially private small datasets. DPRP has several key advantages over the state-of-the-art. Using seven diverse real-life datasets, we show that DPRP outperforms the current state-of-the-art on a variety of tasks, under varying conditions, and for all privacy budgets.}
}