Post-Average Adversarial Defense

Implementation of the Post-Average adversarial defense method as described in Bandlimiting Neural Networks Against Adversarial Attacks.

This implementation is based on PyTorch and uses the Foolbox toolbox to provide attacking methods.

robustml evaluation

This implementation supports the robustml API for evaluation.

To evaluate on CIFAR-10:

python robustml_test_cifar10.py <datasetPath>

To evaluate on ImageNet:

python robustml_test_imagenet.py <datasetPath>