Asymmetric Differential Privacy

TL;DR

This paper introduces asymmetric differential privacy (ADP), a relaxation of standard differential privacy that aims to provide one-sided error guarantees, improving utility for epidemic analysis tasks.

Contribution

The paper proposes ADP, a new privacy framework that allows for one-sided error, addressing limitations of traditional DP in epidemic data analysis.

Findings

ADP achieves one-sided error in privacy-preserving data publication.

Experiments demonstrate ADP's practicality and utility in epidemic analysis.

ADP balances privacy protection with improved data utility.

Abstract

Differential privacy (DP) is getting attention as a privacy definition when publishing statistics of a dataset. This paper focuses on the limitation that DP inevitably causes two-sided error, which is not desirable for epidemic analysis such as how many COVID-19 infected individuals visited location A. For example, consider publishing misinformation that many infected people did not visit location A, which may lead to miss decision-making that expands the epidemic. To fix this issue, we propose a relaxation of DP, called asymmetric differential privacy (ADP). We show that ADP can provide reasonable privacy protection while achieving one-sided error. Finally, we conduct experiments to evaluate the utility of proposed mechanisms for epidemic analysis using a real-world dataset, which shows the practicality of our mechanisms.