Adversarial Examples Are Not Bugs, They Are Features
Abstract:
Adversarial examples have attracted significant attention in machine learning, but the reasons for their existence and pervasiveness remain unclear. We demonstrate that adversarial examples can be directly attributed to the presence of non-robust features: features derived from patterns in the data distribution that are highly predictive, yet brittle and incomprehensible to humans. After capturing these features within a theoretical framework, we establish their widespread existence in standard datasets. Finally, we present a simple setting where we can rigorously tie the phenomena we observe in practice to a misalignment between the (human-specified) notion of robustness and the inherent geometry of the data.
Authors: Andrew Ilyas, Shibani Santurkar, Dimitris Tsipras, Logan Engstrom, Brandon Tran, Aleksander Madry
https://arxiv.org/abs/1905.02175
Смотрите видео Adversarial Examples Are Not Bugs, They Are Features онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Yannic Kilcher 14 Май 2019, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 11,916 раз и оно понравилось 400 людям.