Adversarial poisoning and inverse poisoning against deep learning - Archive ouverte HAL
Pré-Publication, Document De Travail Année : 2019

Adversarial poisoning and inverse poisoning against deep learning

Résumé

Efficient attacks for both adversarial poisoning and adversarial inverse poisoning have recently been found on frozen deep feature plus support vector machine. But, new experiments show that such attacks only works for poisoning, but, not inverse poisoning when targeting deep networks. Investigating this observation, this paper found that adversarial poisoning attacks based on energetic landscape modification outperforms previous ones based on energetic minimum modification. This way, this paper shows that stochastic training is not sufficient defence against inverse poisoning.
Fichier principal
Vignette du fichier
main.pdf (249.53 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02139074 , version 1 (24-05-2019)
hal-02139074 , version 2 (23-06-2019)
hal-02139074 , version 3 (27-05-2021)
hal-02139074 , version 4 (03-11-2021)
hal-02139074 , version 5 (25-09-2024)

Identifiants

  • HAL Id : hal-02139074 , version 1

Citer

Adrien Chan-Hon-Tong. Adversarial poisoning and inverse poisoning against deep learning. 2019. ⟨hal-02139074v1⟩
129 Consultations
497 Téléchargements

Partager

More