Multitask Learning Strengthens Adversarial Robustness

Although deep networks achieve strong accuracy on a range of computer vision benchmarks, they remain vulnerable to adversarial attacks, where imperceptible input perturbations fool the network. We present both theoretical and empirical analyses that connect the adversarial robustness of a model to t...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2020-09
Hauptverfasser:	Mao, Chengzhi, Gupta, Amogh, Vikram Nitin, Ray, Baishakhi, Song, Shuran, Yang, Junfeng, Vondrick, Carl
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer vision Empirical analysis Robustness
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!