Limitations of the Lipschitz constant as a defense against adversarial examples

Authors: Todd Huster, Cho-Yu Jason Chiang, Ritu Chadha
Published: 2018-07-25

Source: https://arxiv.org/abs/1807.09705

PDF: https://arxiv.org/pdf/1807.09705

Labels Predicted by AI

敵対的サンプル ロバスト性に関する評価 モデル評価

Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.

Abstract

Several recent papers have discussed utilizing Lipschitz constants to limit the susceptibility of neural networks to adversarial examples. We analyze recently proposed methods for computing the Lipschitz constant. We show that the Lipschitz constant may indeed enable adversarially robust neural networks. However, the methods currently employed for computing it suffer from theoretical and practical limitations. We argue that addressing this shortcoming is a promising direction for future research into certified adversarial defenses.