Harnessing adversarial examples with a surprisingly simple defense
This is an incremental improvement for enhancing robustness in machine learning models against adversarial attacks.
The paper tackles the problem of defending against adversarial examples by proposing a simple method that adjusts the slope of the ReLU function at test time, showing effectiveness on MNIST and CIFAR-10 datasets against various attacks.
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.