Defense Against Adversarial Attacks via Controlling Gradient Leaking on Embedded Manifolds
Authors:
- 3 November 2020
- book chapter
- Published by Springer Nature
- p. 753-769
Abstract
No abstract availableKeywords
This publication has 12 references indexed in Scilit:
- Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness Against Adversarial Attack2019
- Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses2019
- Tangent-Normal Adversarial Regularization for Semi-Supervised Learning2019
- Improving DNN Robustness to Adversarial Attacks Using Jacobian Regularization2018
- Empirical Study of the Topology and Geometry of Deep Networks2018
- Defense Against Adversarial Attacks Using High-Level Representation Guided Denoiser2018
- Densely Connected Convolutional Networks2017
- Towards Evaluating the Robustness of Neural Networks2017
- Wide Residual Networks2016
- Large-scale malware classification using random projections and neural networks2013