Harnessing adversarial examples with a surprisingly simple defense
arXiv:2004.13013
Abstract
I introduce a very simple method to defend against adversarial examples. The basic idea is to raise the slope of the ReLU function at the test time. Experiments over MNIST and CIFAR-10 datasets demonstrate the effectiveness of the proposed defense against a number of strong attacks in both untargeted and targeted settings. While perhaps not as effective as the state of the art adversarial defenses, this approach can provide insights to understand and mitigate adversarial attacks. It can also be used in conjunction with other defenses.
References in corpus (7)
- Ensemble Adversarial Training: Attacks and Defenses
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
- PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
- Stochastic Activation Pruning for Robust Adversarial Defense
- Mitigating Adversarial Effects Through Randomization
- White Noise Analysis of Neural Networks