Here's our new framework for Privacy Attacks in Federated Learning

Follow the full discussion on Reddit.
We recently uploaded our framework for privacy attacks in federated learning. You can find it here: https://github.com/JonasGeiping/breaching. We include a sizable number of known attacks for deep neural networks in vision and text domains and some new ones.