Learning to Protect Communications with Adversarial Neural Cryptography

Martín Abadi, David G. Andersen

Oct 21, 2016 (modified: Oct 21, 2016) ICLR 2017 conference submission readers: everyone
  • Abstract: We ask whether neural networks can learn to use secret keys to protect information from other neural networks. Specifically, we focus on ensuring confidentiality properties in a multiagent system, and we specify those properties in terms of an adversary. Thus, a system may consist of neural networks named Alice and Bob, and we aim to limit what a third neural network named Eve learns from eavesdropping on the communication between Alice and Bob. We do not prescribe specific cryptographic algorithms to these neural networks; instead, we train end-to-end, adversarially. We demonstrate that the neural networks can learn how to perform forms of encryption and decryption, and also how to apply these operations selectively in order to meet confidentiality goals.
  • TL;DR: Adversarial training of neural networks to learn rudimentary forms of encryption with no pre-specified algorithms
  • Conflicts: google.com, cmu.edu