Adversarial anchor-guided feature refinement for adversarial defense

Cited 2 time in webofscience Cited 0 time in scopus
  • Hit : 158
  • Download : 0
Adversarial training (AT), which is known as a robust training method for defending against adversarial examples, usually loses the performance of models for clean examples due to the feature distribution discrepancy between clean and adversarial. In this paper, we propose a novel Adversarial Anchor-guided Feature Refinement (AAFR) defense method aimed at reducing the discrepancy and delivering reliable performances for both clean and adversarial examples. We devise adversarial anchor that detects whether the feature comes from clean or adversarial example. Then, we use adversarial anchor to refine the feature to reduce the discrepancy. As a result, the proposed method substantially achieves adversarial robustness while preserving the performance for clean examples. The effectiveness of the proposed method is verified with comprehensive experiments on CIFAR-10, CIFAR-100, and Tiny ImageNet datasets.
Publisher
ELSEVIER
Issue Date
2023-08
Language
English
Article Type
Article
Citation

IMAGE AND VISION COMPUTING, v.136

ISSN
0262-8856
DOI
10.1016/j.imavis.2023.104722
URI
http://hdl.handle.net/10203/310051
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 2 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0