The Test of Tests: A Framework for Differentially Private Hypothesis Testing

Zeki Kazan, Kaiyan Shi, Adam Groce, Andrew P Bray
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:16131-16151, 2023.

Abstract

We present a generic framework for creating differentially private versions of any hypothesis test in a black-box way. We analyze the resulting tests analytically and experimentally. Most crucially, we show good practical performance for small data sets, showing that at ε = 1 we only need 5-6 times as much data as in the fully public setting. We compare our work to the one existing framework of this type, as well as to several individually-designed private hypothesis tests. Our framework is higher power than other generic solutions and at least competitive with (and often better than) individually-designed tests.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-kazan23a, title = {The Test of Tests: A Framework for Differentially Private Hypothesis Testing}, author = {Kazan, Zeki and Shi, Kaiyan and Groce, Adam and Bray, Andrew P}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {16131--16151}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/kazan23a/kazan23a.pdf}, url = {https://proceedings.mlr.press/v202/kazan23a.html}, abstract = {We present a generic framework for creating differentially private versions of any hypothesis test in a black-box way. We analyze the resulting tests analytically and experimentally. Most crucially, we show good practical performance for small data sets, showing that at ε = 1 we only need 5-6 times as much data as in the fully public setting. We compare our work to the one existing framework of this type, as well as to several individually-designed private hypothesis tests. Our framework is higher power than other generic solutions and at least competitive with (and often better than) individually-designed tests.} }
Endnote
%0 Conference Paper %T The Test of Tests: A Framework for Differentially Private Hypothesis Testing %A Zeki Kazan %A Kaiyan Shi %A Adam Groce %A Andrew P Bray %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-kazan23a %I PMLR %P 16131--16151 %U https://proceedings.mlr.press/v202/kazan23a.html %V 202 %X We present a generic framework for creating differentially private versions of any hypothesis test in a black-box way. We analyze the resulting tests analytically and experimentally. Most crucially, we show good practical performance for small data sets, showing that at ε = 1 we only need 5-6 times as much data as in the fully public setting. We compare our work to the one existing framework of this type, as well as to several individually-designed private hypothesis tests. Our framework is higher power than other generic solutions and at least competitive with (and often better than) individually-designed tests.
APA
Kazan, Z., Shi, K., Groce, A. & Bray, A.P.. (2023). The Test of Tests: A Framework for Differentially Private Hypothesis Testing. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:16131-16151 Available from https://proceedings.mlr.press/v202/kazan23a.html.

Related Material