Adversarially Robust Learning with Unknown Perturbation Sets

Authors: Omar Montasser, Steve Hanneke, Nathan Srebro
Published: 2021-02-03

Source: https://arxiv.org/abs/2102.02145

PDF: https://arxiv.org/pdf/2102.02145

AIにより推定されたラベル

敵対的サンプルの検知 モデル性能評価 敵対的訓練

※ 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

We study the problem of learning predictors that are robust to adversarial examples with respect to an unknown perturbation set, relying instead on interaction with an adversarial attacker or access to attack oracles, examining different models for such interactions. We obtain upper bounds on the sample complexity and upper and lower bounds on the number of required interactions, or number of successful attacks, in different interaction models, in terms of the VC and Littlestone dimensions of the hypothesis class of predictors, and without any assumptions on the perturbation set.