Learning Private Representations through Entropy-based Adversarial Training

TL;DR

This paper introduces a novel adversarial training method using focal entropy to learn data representations that balance high predictive utility with user privacy preservation, demonstrated across multiple benchmarks.

Contribution

It proposes a new entropy variant, focal entropy, to improve privacy-utility trade-offs in adversarial representation learning.

Findings

High target utility achieved with moderate privacy leakage

Focal entropy reduces information leakage compared to existing methods

Effective across multiple benchmark datasets

Abstract

How can we learn a representation with high predictive power while preserving user privacy? We present an adversarial representation learning method for sanitizing sensitive content from the learned representation. Specifically, we introduce a variant of entropy - focal entropy, which mitigates the potential information leakage of the existing entropy-based approaches. We showcase feasibility on multiple benchmarks. The results suggest high target utility at moderate privacy leakage.