Is My Data in Your AI? Membership Inference Test (MINT) applied to Face Biometrics

TL;DR

This paper presents MINT, a new method to empirically determine if specific data was used in training face recognition AI models, achieving up to 90% accuracy and aiding privacy enforcement.

Contribution

The paper introduces MINT, a novel architecture using MLPs and CNNs to detect training data membership in face recognition models, with extensive experiments on large datasets.

Findings

MINT achieves up to 90% accuracy in membership inference.

The approach is effective across multiple face recognition systems.

It can help enforce privacy and fairness in AI applications.

Abstract

This article introduces the Membership Inference Test (MINT), a novel approach that aims to empirically assess if given data was used during the training of AI/ML models. Specifically, we propose two MINT architectures designed to learn the distinct activation patterns that emerge when an Audited Model is exposed to data used during its training process. These architectures are based on Multilayer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs). The experimental framework focuses on the challenging task of Face Recognition, considering three state-of-the-art Face Recognition systems. Experiments are carried out using six publicly available databases, comprising over 22 million face images in total. Different experimental scenarios are considered depending on the context of the AI model to test. Our proposed MINT approach achieves promising results, with up to 90\% accuracy, indicating the potential to recognize if an AI model has been trained with specific data. The proposed MINT approach can serve to enforce privacy and fairness in several AI applications, e.g., revealing if sensitive or private data was used for training or tuning Large Language Models (LLMs).