Studying the Transferability of Non-Targeted Adversarial Attacks
Empreu sempre aquest identificador per citar o enllaçar aquest ítem
http://hdl.handle.net/10045/138459
Títol: | Studying the Transferability of Non-Targeted Adversarial Attacks |
---|---|
Autors: | Álvarez, Enrique | Alvarez, Rafael | Cazorla, Miguel |
Grups d'investigació o GITE: | Criptología y Seguridad Computacional | Robótica y Visión Tridimensional (RoViT) |
Centre, Departament o Servei: | Universidad de Alicante. Departamento de Ciencia de la Computación e Inteligencia Artificial |
Paraules clau: | Deep Learning | Adversarial Attacks | Convolutional Neural Networks |
Data de publicació: | 22-de setembre-2021 |
Editor: | IEEE |
Citació bibliogràfica: | E. Álvarez, R. Álvarez and M. Cazorla, "Studying the Transferability of Non-Targeted Adversarial Attacks," 2021 International Joint Conference on Neural Networks (IJCNN), Shenzhen, China, 2021, pp. 1-6, doi: 10.1109/IJCNN52387.2021.9534138 |
Resum: | There is no doubt that the use of machine learning is increasing every day. Its applications include self-driving cars, malware detection, recommendation systems and many other fields. Although the broad scope of this technology highlights the importance of its reliability, it has been shown that machine learning models can be vulnerable to adversarial attacks. In this paper, we study a property of these attacks called transferability across different architectures and models, measuring how these attacks transfer based on a specific number of parameters among three adversarial attacks: Fast Gradient Sign Method, Projected Gradient Descent and HopSkipJumpAttack. |
Patrocinadors: | Experiments were made possible by a generous hardware donation from NVIDIA. Research partially supported by the Spanish Government under project grant RTI2018-097263-B-I00 (ACTIS). |
URI: | http://hdl.handle.net/10045/138459 |
ISBN: | 978-1-6654-3900-8 |
ISSN: | 2161-4407 |
DOI: | 10.1109/IJCNN52387.2021.9534138 |
Idioma: | spa |
Tipus: | info:eu-repo/semantics/conferenceObject |
Drets: | © IEEE |
Revisió científica: | si |
Versió de l'editor: | https://doi.org/10.1109/IJCNN52387.2021.9534138 |
Apareix a la col·lecció: | INV - CSC - Comunicaciones a Congresos, Conferencias, etc. INV - RoViT - Comunicaciones a Congresos, Conferencias, etc. |
Arxius per aquest ítem:
Arxiu | Descripció | Tamany | Format | |
---|---|---|---|---|
Studying_the_Transferability_of_Non-Targeted_Adversarial_Attacks.pdf | Versión final (acceso restringido) | 1,11 MB | Adobe PDF | Obrir Sol·licitar una còpia |
Studying_the_Transferability_of_Non-Targeted_Adversarial_Attacks-rev.pdf | Versión revisada (acceso abierto) | 2,78 MB | Adobe PDF | Obrir Vista prèvia |
Tots els documents dipositats a RUA estan protegits per drets d'autors. Alguns drets reservats.