Privacy-Aware Randomized Quantization via Linear Programming

TL;DR

This paper introduces a family of unbiased, differentially private quantization mechanisms formulated via linear programming, improving the privacy-accuracy trade-off for discrete data outputs.

Contribution

It proposes a novel, flexible quantization mechanism family that is unbiased and differentially private, unifying existing methods as special cases, and optimizes the trade-off using linear programming.

Findings

Achieves better privacy-accuracy trade-off than baseline methods

Unbiased and differentially private quantization mechanisms

Formulated as an efficiently solvable linear optimization problem

Abstract

Differential privacy mechanisms such as the Gaussian or Laplace mechanism have been widely used in data analytics for preserving individual privacy. However, they are mostly designed for continuous outputs and are unsuitable for scenarios where discrete values are necessary. Although various quantization mechanisms were proposed recently to generate discrete outputs under differential privacy, the outcomes are either biased or have an inferior accuracy-privacy trade-off. In this paper, we propose a family of quantization mechanisms that is unbiased and differentially private. It has a high degree of freedom and we show that some existing mechanisms can be considered as special cases of ours. To find the optimal mechanism, we formulate a linear optimization that can be solved efficiently using linear programming tools. Experiments show that our proposed mechanism can attain a better privacy-accuracy trade-off compared to baselines.