Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses

Yingwei Li; Song Bai; Cihang Xie; Zhenyu Liao; Xiaohui Shen; Alan Yuille

ECCV 2020poster71 citations

Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses

Yingwei Li, Song Bai, Cihang Xie, Zhenyu Liao, Xiaohui Shen, Alan Yuille

Abstract

This paper focuses on learning transferable adversarial examples specifically against defense models (models to defense adversarial attacks). In particular, we show that a simple universal perturbation can fool a series of state-of-the-art defenses.

BibTeX

@inproceedings{eccv2020_regionalhomogene,
  title = {Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses},
  author = {Yingwei Li and Song Bai and Cihang Xie and Zhenyu Liao and Xiaohui Shen and Alan Yuille},
  booktitle = {ECCV 2020},
  year = {2020}
}