Learning to Noise: Application-Agnostic Data Sharing with Local Differential Privacy

Alex Mansbridge, Gregory Barbour, Davide Piras, Christopher Frye, Ilya Feige, David Barber

Keywords: Differential Privacy, Representation Learning, Variational Inference, Generative Modelling

Abstract: In recent years, the collection and sharing of individuals’ private data has become commonplace in many industries. Local differential privacy (LDP) is a rigorous approach which uses a randomized algorithm to preserve privacy even from the database administrator, unlike the more standard central differential privacy. For LDP, when applying noise directly to high-dimensional data, the level of noise required all but entirely destroys data utility. In this paper we introduce a novel, application-agnostic privatization mechanism that leverages representation learning to overcome the prohibitive noise requirements of direct methods, while maintaining the strict guarantees of LDP. We further demonstrate that data privatized with this mechanism can be used to train machine learning algorithms. Applications of this model include private data collection, private novel-class classification, and the augmentation of clean datasets with additional privatized features. We achieve significant gains in performance on downstream classification tasks relative to benchmarks that noise the data directly, which are state-of-the-art in the context of application-agnostic LDP mechanisms for high-dimensional data sharing tasks.

One-sentence Summary: Using representation learning to induce local differential privacy on high-dimensional data, via an application-agnostic privatization mechanism.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=Hp1ZC6tDeB