Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters

Archive ouverte : Communication dans un congrès

El-Allami, R. | Marchisio, A. | Shafique, M. | Alouani, Ihsen

Edité par HAL CCSD ; IEEE

International audience. Deep Learning (DL) algorithms have gained popularity owing to their practical problem-solving capacity. However, they suffer from a serious integrity threat, i.e., their vulnerability to adversarial attacks. In the quest for DL trustworthiness, recent works claimed the inherent robustness of Spiking Neural Networks (SNNs) to these attacks, without considering the variability in their structural spiking parameters. This paper explores the security enhancement of SNNs through internal structural parameters. Specifically, we investigate the SNNs robustness to adversarial attacks with different values of the neuron's firing voltage thresholds and time window boundaries. We thoroughly study SNNs security under different adversarial attacks in the strong white-box setting, with different noise budgets and under variable spiking parameters. Our results show a significant impact of the structural parameters on the SNNs' security, and promising sweet spots can be reached to design trustworthy SNNs with 85% higher robustness than a traditional non-spiking DL system. To the best of our knowledge, this is the first work that investigates the impact of structural parameters on SNNs robustness to adversarial attacks. The proposed contributions and the experimental framework is available online 11https://github.com/rda-ela/SNN-Adversarial-Attacks to the community for reproducible research. © 2021 EDAA.

Consulter en ligne

Suggestions

Du même auteur

CNTFET-Based Ternary Multiply-and-Accumulate Unit

Archive ouverte: Article de revue

Mohammaden, Amr | 2022

International audience. Multiply-Accumulate (MAC) is one of the most commonly used operations in modern computing systems due to its use in matrix multiplication, signal processing, and in new applications such as m...

Adversarial Attacks in a Multi-view Setting: An Empirical Study of the Adve...

Archive ouverte: Communication dans un congrès

Tarchoun, Bilel | 2021-09-28

International audience. While machine learning applications are getting mainstream owing to a demonstrated efficiency in solving complex problems, they suffer from inherent vulnerability to adversarial attacks. Adve...

Enabling User Relaying In MCM-NOMA Under Doubly Selective Channels Using It...

Archive ouverte: Article de revue

Hamza, Anis Amazigh | 2022-03

International audience. Cell-edge users of future cellular internet of things (IoT) sensor networks can suffer from extremely severe channel conditions, especially under mobility with high speeds, i.e., doubly selec...

Du même sujet

Jours de sable / Aimée de Jongh

Livre | Jongh, Aimée de (1988-....). Auteur. Illustrateur | 2021

Washington, 1937. John Clark, journaliste photoreporter de 22 ans, est engagé par la Farm Security Administration, l'organisme gouvernemental chargé d'aider les fermiers victimes de la Grande Dépression. Sa mission : témoigner de ...

Lower Voltage for Higher Security: Using Voltage Overscaling to Secure Deep...

Archive ouverte: Communication dans un congrès

Islam, Shohidul | 2021-11-01

International audience. Deep neural networks (DNNs) are shown to be vulnerable to adversarial attacks-- carefully crafted additive noise that undermines DNNs integrity. Previously proposed defenses against these att...

Saveurs du récit biblique / Daniel Marguerat, André Wénin

Livre | Marguerat, Daniel (1943-....). Auteur | 2012

L' investisseur intelligent : une référence en matière d'investissement / B...

Livre | Graham, Benjamin (1894-1976). Auteur | 2018 - [Nouvelle édition commentée]

"Benjamin Graham, le plus grand financier du XXème siècle, a inspiré nombre d'investisseurs individuels et professionnels. Sa stratégie - l'investissement dans la valeur - permet de bâtir des stratégies de long terme et d'éviter d...

Fundamentals of strategy / Gerry Johnson.., Richard Whittington,.. Kevan Sc...

Livre | Johnson, Gerry (19..-....). Auteur | 2012 - Second edition

Visualisation et analyse statistique de données : concepts de base, tableur...

Livre | Seigneuric, Renaud (19..-....) - auteur en ingéniérie pour la santé. Auteur | 2022

"Cet ouvrage est le fruit de plus de 20 années combinant recherche, pratique et enseignements en statistiques et analyse de données. Il vise à rendre rapides, concrètes, interactives et finalement stimulantes les statistiques et l...

Chargement des enrichissements...