Adversarial Robustness is a growing field that evidences the brittleness of
neural networks. Although the literature on adversarial robustness is vast, a
dimension is missing in these studies: assessing how severe the mistakes are.
We call this notion "Adversarial Severity" since it quantifies the downstream
impact of adversarial corruptions by computing the semantic error between the
misclassification and the proper label. We propose to study the effects of
adversarial noise by measuring the Robustness and Severity into a large-scale
dataset: iNaturalist-H. Our contributions are: (i) we introduce novel
Hierarchical Attacks that harness the rich structured space of labels to create
adversarial examples. (ii) These attacks allow us to benchmark the Adversarial
Robustness and Severity of classification models. (iii) We enhance the
traditional adversarial training with a simple yet effective Hierarchical
Curriculum Training to learn these nodes gradually within the hierarchical
tree. We perform extensive experiments showing that hierarchical defenses allow
deep models to boost the adversarial Robustness by 1.85% and reduce the
severity of all attacks by 0.17, on average.

本文介绍了在 Adversarial Robustness 领域中，量化 adversarial noise 的影响的新概念 ——Adversarial Severity，并提出了使用基于 Hierarchical Attacks 的新型防护方法和 Hierarchical Curriculum Training 训练方法来提高深度模型的 Robustness 和减轻所有攻击的 severity。实验表明，这种防护方法可以使深度模型的 Robustness 提高 1.85％，平均减少攻击的 severity 0.17。