Learning to Protect Communications with Adversarial Neural Cryptography
This addresses the challenge of secure communication in multiagent AI systems, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of enabling neural networks to learn encryption and decryption to protect communications from eavesdropping by other neural networks, demonstrating that they can achieve confidentiality goals through adversarial training without prescribed algorithms.
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.