中文

利用集成自适应梯度生成最小对抗扰动

密码学与安全 2019-05-21 v3 机器学习 机器学习

摘要

深度神经网络容易被对抗样本以高置信度预测所欺骗。

关键词

引用

@article{arxiv.1904.06186,
  title  = {Generating Minimal Adversarial Perturbations with Integrated Adaptive Gradients},
  author = {Yatie Xiao and Chi-Man Pun},
  journal= {arXiv preprint arXiv:1904.06186},
  year   = {2019}
}

备注

8 pages, 7 figures. arXiv admin note: text overlap with arXiv:1902.01220 The formula in Algorithm 1 lacks important representations