EUVIMEDEUVIMED entdecken
Crossref

Adaptive Adversarial Logits Pairing

Shangxi Wu, Jitao Sang, Kaiyan Xu, Guanhua Zheng, Changsheng Xu

ACM Transactions on Multimedia Computing, Communications, and Applications · 29.02.2024 · 20 (2) · 1-16

DOI: 10.1145/3616375 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Multimedia Computing, Communications, and Applications, online erschienen am 18.10.2023. Verfasst von Shangxi Wu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: Fundamental Research Funds for the Central Universities, National Natural Science Foundation of China, Beijing Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Adversarial examples provide an opportunity as well as impose a challenge for understanding image classification systems. Based on the analysis of the adversarial training solution—Adversarial Logits Pairing (ALP), we observed in this work that: (1) The inference of adversarially robust model tends to rely on fewer high-contribution features compared with vulnerable ones. (2) The training target of ALP does not fit well to a noticeable part of samples, where the logits pairing loss is overemphasized and obstructs minimizing the classification loss. Motivated by these observations, we design an Adaptive Adversarial Logits Pairing (AALP) solution by modifying the training process and training target of ALP. Specifically, AALP consists of an adaptive feature optimization module with Guided Dropout to systematically pursue fewer high-contribution features, and an adaptive sample weighting module by setting sample-specific training weights to balance between logits pairing loss and classification loss. The proposed AALP solution demonstrates superior defense performance on multiple datasets with extensive experiments.

Förderung

  • Fundamental Research Funds for the Central Universities
  • National Natural Science Foundation of China
  • Beijing Natural Science Foundation
  • CCF-Zhipu AI Large Model Fund

Zitieren

APA

Wu, S., Sang, J., Xu, K., Zheng, G., & Xu, C. (2024). Adaptive Adversarial Logits Pairing. ACM Transactions on Multimedia Computing, Communications, and Applications, 20(2), 1-16. https://doi.org/10.1145/3616375

Vancouver

Wu S, Sang J, Xu K, Zheng G, Xu C. Adaptive Adversarial Logits Pairing. ACM Transactions on Multimedia Computing, Communications, and Applications. 2024;20(2):1-16. doi:10.1145/3616375

BibTeX
@article{Wu2024,
  author = {Wu, Shangxi and Sang, Jitao and Xu, Kaiyan and Zheng, Guanhua and Xu, Changsheng},
  title = {Adaptive Adversarial Logits Pairing},
  journal = {ACM Transactions on Multimedia Computing, Communications, and Applications},
  year = {2024},
  volume = {20},
  number = {2},
  pages = {1--16},
  doi = {10.1145/3616375}
}
Kostenlos starten →