BriefGPT.xyz
Jun, 2022
认证的免费对抗鲁棒性
(Certified!!) Adversarial Robustness for Free!
HTML
PDF
Nicholas Carlini, Florian Tramer, Krishnamurthy, Dvijotham, J. Zico Kolter
TL;DR
本文介绍了一个基于离线预训练模型,通过组合去噪扩散概率模型和高性能分类器等手段实现了对于2-范数边界扰动的认证敌对鲁棒性,并在ImageNet数据集上得到了71%的分类准确率,显著优于之前的相关研究。
Abstract
In this paper we show how to achieve state-of-the-art
certified adversarial robustness
to
2-norm bounded perturbations
by relying exclusively on off-the-shelf
→