Improving Frequency Estimation under Local Differential Privacy

TL;DR

This paper advances the understanding of frequency estimation under local differential privacy by deriving tighter bounds using information theory and proposing improved estimators that outperform existing methods.

Contribution

It introduces new information-theoretic bounds for frequency estimation under local differential privacy and develops estimators that achieve these bounds for binary data.

Findings

New tighter bounds on the privacy-utility tradeoff.

Proposed estimators outperform state-of-the-art methods.

Bounds are attainable for binary inputs.

Abstract

Local Differential Privacy protocols are stochastic protocols used in data aggregation when individual users do not trust the data aggregator with their private data. In such protocols there is a fundamental tradeoff between user privacy and aggregator utility. In the setting of frequency estimation, established bounds on this tradeoff are either nonquantitative, or far from what is known to be attainable. In this paper, we use information-theoretical methods to significantly improve established bounds. We also show that the new bounds are attainable for binary inputs. Furthermore, our methods lead to improved frequency estimators, which we experimentally show to outperform state-of-the-art methods.