Adversarial Neural Cryptography
VISHAL KIRAN (~vishal75) |
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.
In this talk, I demonstrate that the neural networks can learn how to perform forms of encryption and decryption using Python Libraries like Theano, and also how to apply these operations selectively in order to meet confidentiality goals. Basically, this talk will make the audience aware of how we can use Python, Machine Learning and Cryptography in tandem to achieve security of communications.
The intended audience will include all data scientists and security researchers. This will help them to achieve an amalgamation of Deep Learning and Cryptography to create strong unbreakable channel to have a secure communication.
Python; Theano; Cryptography basics
A final year student studying at IIIT Vadodara. Machine learning enthusiast. An avid Python user.