Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better

Adversarial training is one effective approach for training robust deep neural networks against adversarial attacks. While being able to bring reliable robustness, adversarial training (AT) methods in general favor high capacity models, i.e., the larger the model the better the robustness. This tend...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2021-08
Hauptverfasser:	Bojia Zi, Zhao, Shihao, Ma, Xingjun, Yu-Gang, Jiang
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial neural networks Distillation Electronic devices Labels Machine learning Robustness Training
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!