A Theoretically Grounded Extension of Universal Attacks from the Attacker's Viewpoint - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

A Theoretically Grounded Extension of Universal Attacks from the Attacker's Viewpoint

Résumé

We extend universal attacks by jointly learning a set of perturbations to choose from to maximize the chance of attacking deep neural network models. Specifically, we embrace the attacker's perspective and introduce a theoretical bound quantifying how much the universal perturbations are able to fool a given model on unseen examples. An extension to assert the transferability of universal attacks is also provided. To learn such perturbations, we devise an algorithmic solution with convergence guarantees under Lipschitz continuity assumptions. Moreover, we demonstrate how it can improve the performance of state-of-the-art gradient-based universal perturbation. As evidenced by our experiments, these novel universal perturbations result in more interpretable, diverse, and transferable attacks.
Fichier principal
Vignette du fichier
ECML_2024___Semi_Universal_Perturbations-1.pdf (822.84 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
licence

Dates et versions

hal-03615461 , version 1 (21-03-2022)
hal-03615461 , version 2 (07-06-2023)
hal-03615461 , version 3 (07-06-2024)

Licence

Identifiants

  • HAL Id : hal-03615461 , version 3

Citer

Jordan Patracone, Paul Viallard, Emilie Morvant, Gilles Gasso, Amaury Habrard, et al.. A Theoretically Grounded Extension of Universal Attacks from the Attacker's Viewpoint. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2024, Vilnius, Lithuania. ⟨hal-03615461v3⟩
133 Consultations
174 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More