On Norm-Agnostic Robustness of Adversarial Training

Authors: Bai Li, Changyou Chen, Wenlin Wang, Lawrence Carin
Published: 2019-05-15

Source: https://arxiv.org/abs/1905.06455

PDF: https://arxiv.org/pdf/1905.06455

AIにより推定されたラベル

ポイズニング 敵対的訓練 敵対的サンプル

※ こちらのラベルはAIによって自動的に追加されました。そのため、正確でないことがあります。
詳細は文献データベースについてをご覧ください。

Abstract

Adversarial examples are carefully perturbed in-puts for fooling machine learning models. A well-acknowledged defense method against such examples is adversarial training, where adversarial examples are injected into training data to increase robustness. In this paper, we propose a new attack to unveil an undesired property of the state-of-the-art adversarial training, that is it fails to obtain robustness against perturbations in ℓ₂ and ℓ_∞ norms simultaneously. We discuss a possible solution to this issue and its limitations as well.