Towards Deep Learning Models Resistant to Adversarial Attacks(PGD) 文章目录 Towards Deep Learning Models Resistant to Adversarial Attacks(PGD)摘要1 介绍2 对抗鲁棒性的优化视角3 面向通用的鲁棒网络3.1 对抗样本的前景…
目录 Introduction内容简介 An Optimization View on Adversarial Robustness内容介绍 Towards Universally Robust NetworksThe Landscape of Adversarial Examples Network Capacity and Adversarial Robustness内容介绍 Experiments: Adversarially Robust Deep Learning Mo…
Towards Deep Learning Models Resistant to Adversarial Attacks
Towards Deep Learning Models Resistant to Adversarial Attacks (PGD),ICLR2018,涉及PGD和对抗训练。
Abstract:本文从优化的角度研究了神经网络的对抗鲁棒性问题。本文提出的方法提…