BriefGPT.xyz
Feb, 2020
广义Lipschitz正则化等于分布式健壮性
Generalised Lipschitz Regularisation Equals Distributional Robustness
HTML
PDF
Zac Cranko, Zhan Shi, Xinhua Zhang, Richard Nock, Simon Kornblith
TL;DR
探讨关于对抗性样本的问题,提出一种基于运输成本不确定性集合的正则化方法,实现了对Lipschitz正则化模型的稳健性证明,揭示了对抗学习和分布鲁棒性之间的联系,并提供了对核分类器的Lipschitz正则化方法。
Abstract
The problem of
adversarial examples
has highlighted the need for a theory of
regularisation
that is general enough to apply to exotic function classes, such as universal approximators. In response, we give a very
→