Towards securing machine learning models against membership inference attacks

Ben Hamida, S; Mrabet, H; Belguith, S; Alhomoud, A; Jemai, A

Abstract

From fraud detection to speech recognition, including price prediction,
Machine Learning (ML) applications are manifold and can significantly improve
different areas. Nevertheless, machine learning models are vulnerable and are
exposed to different security and privacy attacks. Hence, these issues should be
addressed while using ML models to preserve the security and privacy of the data
used. There is a need to secure ML models, especially in the training phase to
preserve the privacy of the training datasets and to minimise the information
leakage. In this paper, we present an overview of ML threats and vulnerabilities,
and we highlight current progress in the research works proposing defence
techniques against ML security and privacy attacks. The relevant background for
the different attacks occurring in both the training and testing/inferring phases is
introduced before presenting a detailed overview of Membership Inference
Attacks (MIA) and the related countermeasures. In this paper, we introduce a
countermeasure against membership inference attacks (MIA) on Conventional
Neural Networks (CNN) based on dropout and L2 regularization. Through
experimental analysis, we demonstrate that this defence technique can mitigate the
risks of MIA attacks while ensuring an acceptable accuracy of the model. Indeed,
using CNN model training on two datasets CIFAR-10 and CIFAR-100, we
empirically verify the ability of our defence strategy to decrease the impact of MIA
on our model and we compare results of five different classifiers. Moreover, we
present a solution to achieve a trade-off between the performance of the model and
the mitigation of MIA attack.

• Journal Article Type: Article
• Acceptance Date: Jun 20, 2021
• Online Publication Date: Oct 11, 2021
• Publication Date: Jan 1, 2022
• Deposit Date: Aug 3, 2021
• Publicly Available Date: Nov 1, 2021
• Journal: Computers, Materials & Continua
• Print ISSN: 1546-2218
• Electronic ISSN: 1546-2226
• Publisher: Tech Science Press
• Volume: 70
• Issue: 3
• Pages: 4897-4919
• DOI: https://doi.org/10.32604/cmc.2022.019709
• Publisher URL: https://doi.org/10.32604/cmc.2022.019709
• Related Public URLs: http://www.techscience.com/journal/cmc