Analyzing the Differentially Private Theil-Sen Estimator for Simple Linear Regression
Cryptography and Security
2024-03-13 v2 Applications
Abstract
In this paper, we study differentially private point and confidence interval estimators for simple linear regression. Motivated by recent work that highlights the strong empirical performance of an algorithm based on robust statistics, DPTheilSen, we provide a rigorous, finite-sample analysis of its privacy and accuracy properties, offer guidance on setting hyperparameters, and show how to produce differentially private confidence intervals to accompany its point estimates.
Keywords
Cite
@article{arxiv.2207.13289,
title = {Analyzing the Differentially Private Theil-Sen Estimator for Simple Linear Regression},
author = {Jayshree Sarathy and Salil Vadhan},
journal= {arXiv preprint arXiv:2207.13289},
year = {2024}
}
Comments
Extended abstract presented at the 2021 workshop on Theory and Practice of Differential Privacy