rfratila/Adversarial-Neural-Cryptography

An implementation of the ICLR 2017 paper 'Learning to Protect Communications with Adversarial Neural Cryptography' while extending it in trying new models and testing it against more sophisticated networks.

Python