A Python 2 implementation of Priv'IT: Private and Sample Efficient Identity Testing.
Paper appeared at ICML 2017, arXiv verison available here, authored by Bryan Cai, Constantinos Daskalakis, and Gautam Kamath.
For comparison purposes, contains implementations of Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing, and A New Class of Private Chi-Square Tests.
Code primarily written by Bryan Cai and Gautam Kamath.
If you use our code or paper, we ask that you please cite:
@inproceedings{CaiDK17,
author = {Cai, Bryan and Daskalakis, Constantinos and Kamath, Gautam},
title = {Priv'IT: Private and Sample Efficient Identity Testing},
booktitle = {Proceedings of the 34th International Conference on Machine Learning},
series = {ICML '17},
year = {2017},
pages = {635--644},
publisher = {JMLR, Inc.}
}