Learning to Protect Communications with Adversarial Neural Cryptography

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.

15 pages

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• TensorFlow: A system for large-scale machine learning
• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Improved Techniques for Training GANs
• Learning Multiagent Communication with Backpropagation
• The Variational Fair Autoencoder
• Crypto-Nets: Neural Networks over Encrypted Data
• Censoring Representations with an Adversary
• Neural Programmer: Inducing Latent Programs with Gradient Descent

Cited by in corpus (24)

• NIPS 2016 Tutorial: Generative Adversarial Networks
• Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
• Deep Learning-Based Communication Over the Air
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Deep Learning in Mobile and Wireless Networking: A Survey
• Challenges of AI in Wireless Networks for IoT
• Cooperative Multi-Agent Reinforcement Learning for Low-Level Wireless Communication
• An Introduction to Deep Learning for the Physical Layer
• Message Passing Multi-Agent GANs
• Recent Advances of Image Steganography with Generative Adversarial Networks
• Digital Cardan Grille: A Modern Approach for Information Hiding
• Performing Co-Membership Attacks Against Deep Generative Models
• HiDDeN: Hiding Data With Deep Networks
• Modeling and Simultaneously Removing Bias via Adversarial Neural Networks
• The Reincarnation of Grille Cipher: A Generative Approach
• Distributed generation of privacy preserving data with user customization
• On Computation and Generalization of GANs with Spectrum Control
• Synchronized Detection and Recovery of Steganographic Messages with Adversarial Learning
• Generating private data with user customization
• Generative Steganography by Sampling
• A new neural-network-based model for measuring the strength of a pseudorandom binary sequence
• Using Echo State Networks for Cryptography
• Generative Steganography with Kerckhoffs' Principle
• Loosely Coupled Federated Learning Over Generative Models