BriefGPT.xyz
May, 2018
通过结构化梯度正则化进行对抗性鲁棒性训练
Adversarially Robust Training through Structured Gradient Regularization
HTML
PDF
Kevin Roth, Aurelien Lucchi, Sebastian Nowozin, Thomas Hofmann
TL;DR
本文提出了一种新的数据依赖性结构化梯度正则化器，旨在增加神经网络对抗扰动的鲁棒性，该正则化器可以从第一原理中导出。实验证据表明，结构化梯度正则化是对抗低水平信号污染攻击的有效一线防御。
Abstract
We propose a novel data-dependent
structured gradient regularizer
to increase the robustness of
neural networks
vis-a-vis
adversarial perturbatio
→