ADePT: Auto-encoder based Differentially Private Text Transformation

TL;DR

This paper introduces ADePT, an auto-encoder based method for differentially private text transformation that maintains high semantic quality and improves privacy protection in NLP tasks.

Contribution

It presents a novel auto-encoder based algorithm that preserves utility and enhances privacy in differentially private text transformations for NLP.

Findings

Outperforms existing methods in resisting Membership Inference Attacks

Maintains high utility with minimal degradation in NLP tasks

Provides theoretical privacy guarantees

Abstract

Privacy is an important concern when building statistical models on data containing personal information. Differential privacy offers a strong definition of privacy and can be used to solve several privacy concerns (Dwork et al., 2014). Multiple solutions have been proposed for the differentially-private transformation of datasets containing sensitive information. However, such transformation algorithms offer poor utility in Natural Language Processing (NLP) tasks due to noise added in the process. In this paper, we address this issue by providing a utility-preserving differentially private text transformation algorithm using auto-encoders. Our algorithm transforms text to offer robustness against attacks and produces transformations with high semantic quality that perform well on downstream NLP tasks. We prove the theoretical privacy guarantee of our algorithm and assess its privacy leakage under Membership Inference Attacks(MIA) (Shokri et al., 2017) on models trained with transformed data. Our results show that the proposed model performs better against MIA attacks while offering lower to no degradation in the utility of the underlying transformation process compared to existing baselines.