Learning to Protect Communications with Adversarial Neural Cryptography

@article{Abadi2016LearningTP,
  title={Learning to Protect Communications with Adversarial Neural Cryptography},
  author={Mart{\'i}n Abadi and David G. Andersen},
  journal={CoRR},
  year={2016},
  volume={abs/1610.06918}
}
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… CONTINUE READING
Highly Cited
This paper has 48 citations. REVIEW CITATIONS
Recent Discussions
This paper has been referenced on Twitter 481 times over the past 90 days. VIEW TWEETS

6 Figures & Tables

Topics

Statistics

02040201620172018
Citations per Year

Citation Velocity: 19

Averaging 19 citations per year over the last 3 years.

Learn more about how we calculate this metric in our FAQ.
Blog posts, news articles and tweet counts and IDs sourced by
Altmetric.com