• 2024-09-27李宏毅机器学习2023-HW10-Adversarial Attack
    文章目录TaskBaselineFGSM(FastGradientSignMethod(FGSM)I-FGSM(IterativeFastGradientSignMethod)MI-FGSM(MomentumIterativeFastGradientSignMethod)M-DI2-FGSM(DiverseInputMomentumIterativeFastGradientSignMethod)ReportfgsmattackJepgCom
  • 2024-08-21对抗训练综述学习笔记
    本文基于这篇综述进行讨论学习:《BaiT,LuoJ,ZhaoJ,etal.Recentadvancesinadversarialtrainingforadversarialrobustness[J].arXivpreprintarXiv:2102.01356,2021.》对抗训练主要目的是让模型能够对于对抗样本做出正确的判断。常见的对抗样本的生成方式(或者说