BriefGPT.xyz
Feb, 2017
关于(统计)检测对抗样本的研究
On the (Statistical) Detection of Adversarial Examples
HTML
PDF
Kathrin Grosse, Praveen Manoharan, Nicolas Papernot, Michael Backes, Patrick McDaniel
TL;DR
本文研究如何检测机器学习中的对抗性样本,提出使用统计检验和模型增强的方法来识别对抗性样本,并参照多个数据集和对抗样本制作方法进行实验,结果表明统计学特性对于检测对抗性样本至关重要。
Abstract
machine learning
(ML) models are applied in a variety of tasks such as network intrusion detection or malware classification. Yet, these models are vulnerable to a class of malicious inputs known as
adversarial examples
→