Stochastic Sparse Adversarial Attacks - IRT SystemX Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Stochastic Sparse Adversarial Attacks

Résumé

This paper introduces stochastic sparse adversarial attacks (SSAA), standing as simple, fast and purely noise-based targeted and untargeted attacks of neural network classifiers (NNC). SSAA offer new examples of sparse (or L0) attacks for which only few methods have been proposed previously. These attacks are devised by exploiting a small-time expansion idea widely used for Markov processes. Experiments on small and large datasets (CIFAR-10 and ImageNet) illustrate several advantages of SSAA in comparison with the-state-of-the-art methods. For instance, in the untargeted case, our method called Voting Folded Gaussian Attack (VFGA) scales efficiently to ImageNet and achieves a significantly lower L0 score than SparseFool (up to 2/5 ) while being faster. Moreover, VFGA achieves better L0 scores on ImageNet than Sparse-RS when both attacks are fully successful on a large number of samples.

Dates et versions

Identifiants

Citer

Manon Cesaire, Lucas Schott, Hatem Hajri, Sylvain Lamprier, Patrick Gallinari. Stochastic Sparse Adversarial Attacks. 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI), Nov 2021, Washington, United States. pp.1247-1254, ⟨10.1109/ICTAI52525.2021.00198⟩. ⟨hal-03550949⟩
108 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More