We consider differentially private counting when each data point consists of d bits satisfying a partial order. Our main technical contribution is a problem-specific K-norm mechanism that runs in time O(d2). Experiments show that, depending on the partial order in question, our solution dominates existing pure differentially private mechanisms, and can reduce their error by an order of magnitude or more.
@article{arxiv.2410.06881,
title = {Privately Counting Partially Ordered Data},
author = {Matthew Joseph and Mónica Ribero and Alexander Yu},
journal= {arXiv preprint arXiv:2410.06881},
year = {2024}
}