Robust GCN against adversarial attacks

Hello There,

I have implemented a Robust GCN version using DGL. This is the original paper:

How do I add it to the DGL Library?
What is the process to follow? Please let me know as I am doing this for the first time.
Going forward, I would like to contribute more to the repository.

Another Question:
Do we have any of the adversarial attacks on graph implemented and available in DGL?

Thank you for your interest in making contributions and this sounds like a great work! As far as I know, there isn’t any example in DGL repo for adversarial attacks on graphs. To contribute the example, you need to fork DGL’s repo, implement your model under the examples folder and then open a pull request. You can find bunch of examples here.