ICML 2022spotlight53 citations
FriendlyCore: Practical Differentially Private Aggregation
Eliad Tsfadia, Edith Cohen, Haim Kaplan, Yishay Mansour, Uri Stemmer
Abstract
Differentially private algorithms for common metric aggregation tasks, such as clustering or averaging, often have limited practicality due to their complexity or to the large number of data points that is required for accurate results. We propose a simple and practical tool $\mathsf{FriendlyCore}$ that takes a set of points ${\cal D}$ from an unrestricted (pseudo) metric space as input. When ${\cal D}$ has effective diameter $r$, $\mathsf{FriendlyCore}$ returns a “stable” subset ${\cal C} \subseteq {\cal D}$ that includes all points, except possibly few outliers, and is
BibTeX
@InProceedings{pmlr-v162-tsfadia22a,
title = {{F}riendly{C}ore: Practical Differentially Private Aggregation},
author = {Tsfadia, Eliad and Cohen, Edith and Kaplan, Haim and Mansour, Yishay and Stemmer, Uri},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
pages = {21828--21863},
year = {2022},
editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
volume = {162},
series = {Proceedings of Machine Learning Research},
month = {17--23 Jul},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v162/tsfadia22a/tsfadia22a.pdf},
url = {https://proceedings.mlr.press/v162/tsfadia22a.html},
abstract = {Differentially private algorithms for common metric aggregation tasks, such as clustering or averaging, often have limited practicality due to their complexity or to the large number of data points that is required for accurate results. We propose a simple and practical tool $\mathsf{FriendlyCore}$ that takes a set of points ${\cal D}$ from an unrestricted (pseudo) metric space as input. When ${\cal D}$ has effective diameter $r$, $\mathsf{FriendlyCore}$ returns a “stable” subset ${\cal C} \subseteq {\cal D}$ that includes all points, except possibly few outliers, and is