Practical Adversarial Robustness in Deep Learning: Problems and Solutions (PAR)

By Pin-Yu Chen (IBM Research) and Sayak Paul (Carted)

Contains tutorial notebooks for our tutorial at CVPR 2021. Slides are available here. A pre-recorded video of our tutorial is also available on YouTube here.

Outline

A detailed outline is available on the website.

We provide Jupyter Notebooks to demonstrate some of the topics mentioned in the outline. These notebooks are fully runnable on Google Colab without any non-trivial configurations.

How to run the notebooks?

Here are the steps -

• First, get the Open in Colab Google Chrome extension.
• Follow this screencast that shows how to navigate to a particular notebook inside this repository and open it in Google Colab.

Major libraries used

• TensorFlow
• Keras
• Neural Structured Learning
• Foolbox

Acknowledgements

ML-GDE program for providing Google Cloud Platform credits that were used to conduct our experiments.