Obfuscation via Information Density Estimation

Abstract

Identifying features that leak information about sensitive attributes is a key challenge in the design of information obfuscation mechanisms. In this paper, we propose a framework to identify information-leaking features via information density estimation. Here, features whose information densities exceed a pre-defined threshold are deemed information-leaking features. Once these features are identified, we sequentially pass them through a targeted obfuscation mechanism with a provable leakage guarantee in terms of ${\mathsf{E}}_{\gamma }$-divergence. The core of this mechanism relies on a data-driven estimate of the trimmed information density for which we propose a novel estimator, named the trimmed information density estimator (TIDE). We then use TIDE to implement our mechanism on three real-world datasets. Our approach can be used as a data-driven pipeline for designing obfuscation mechanisms targeting specific features.

24 pages, 3 figures

References in corpus (10)

• Renyi Differential Privacy
• Estimating divergence functionals and the likelihood ratio by convex risk minimization
• Utility-Privacy Tradeoff in Databases: An Information-theoretic Approach
• $f$-divergence Inequalities
• Differential Privacy as a Mutual Information Constraint
• Defeating Image Obfuscation with Deep Learning
• Generative Adversarial Privacy
• Trimmed Density Ratio Estimation
• A Tunable Measure for Information Leakage
• Distributed generation of privacy preserving data with user customization

Cited by in corpus (3)

• Privacy in Deep Learning: A Survey
• Variational Leakage: The Role of Information Complexity in Privacy Leakage
• Uncertainty-Autoencoder-Based Privacy and Utility Preserving Data Type Conscious Transformation