This repository includes several tests that I attempted to implement and execute a membership inference attack.
- Each notebook contains a different approach.
- All of the approaches require the victim model to have overfitted to certain extent on the training data.
TODO:
- Clean this up with proper documentation. (I probably won't)