Privacy Enhanced Multimodal Neural Representations for Emotion Recognition

TL;DR

This paper investigates how multimodal emotion recognition models can unintentionally leak demographic information and proposes an adversarial learning approach to enhance privacy without significantly reducing emotion recognition accuracy.

Contribution

It introduces a novel adversarial training method to unlearn demographic information in multimodal representations for emotion recognition, addressing privacy concerns across different data modalities.

Findings

Adversarial training improves privacy metrics across modalities.

The method maintains primary task performance with minimal accuracy loss.

Different modalities exhibit varying levels of demographic information leakage.

Abstract

Many mobile applications and virtual conversational agents now aim to recognize and adapt to emotions. To enable this, data are transmitted from users' devices and stored on central servers. Yet, these data contain sensitive information that could be used by mobile applications without user's consent or, maliciously, by an eavesdropping adversary. In this work, we show how multimodal representations trained for a primary task, here emotion recognition, can unintentionally leak demographic information, which could override a selected opt-out option by the user. We analyze how this leakage differs in representations obtained from textual, acoustic, and multimodal data. We use an adversarial learning paradigm to unlearn the private information present in a representation and investigate the effect of varying the strength of the adversarial component on the primary task and on the privacy metric, defined here as the inability of an attacker to predict specific demographic information. We evaluate this paradigm on multiple datasets and show that we can improve the privacy metric while not significantly impacting the performance on the primary task. To the best of our knowledge, this is the first work to analyze how the privacy metric differs across modalities and how multiple privacy concerns can be tackled while still maintaining performance on emotion recognition.