https://github.com/amartya18x/tapas
Tricks for Accelerating (encrypted) Prediction As a Service
https://github.com/amartya18x/tapas
binary-neural-networks homomorphic-encryption privacy-preserving-machine-learning
Last synced: 4 months ago
JSON representation
Tricks for Accelerating (encrypted) Prediction As a Service
- Host: GitHub
- URL: https://github.com/amartya18x/tapas
- Owner: amartya18x
- Created: 2018-06-07T21:24:14.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2019-05-28T13:34:49.000Z (about 6 years ago)
- Last Synced: 2025-01-15T13:27:30.908Z (5 months ago)
- Topics: binary-neural-networks, homomorphic-encryption, privacy-preserving-machine-learning
- Language: HTML
- Homepage: https://amartya18x.github.io/tapas
- Size: 2.98 MB
- Stars: 19
- Watchers: 8
- Forks: 6
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
This page provides codes for reproducing our results. We use two main libraries here:
* The very awesome [**SHEEP**](https://github.com/alan-turing-institute/SHEEP) library prepared by[ Oliver and Nick](https://www.turing.ac.uk/research-engineering/#people) to run boolean circuits on ecnrypted data using various homomorphic encryption schemes for our networks.
* Our own general purpose library, **matSHEEP**, which provides a programmatic interface to design and visualize logic circuits for deep neural networks in python. Find more abour **matSHEEP** at [this link](https://amartya18x.github.io/matSHEEP).
There are two main types of experiments.
* [Doing Prediction over Encrypted Data using Binary Neural Netwrks.](https://amartya18x.github.io/tapas/private_predictions)
* [Recording accuracies for various levels of sparsity of a binary neural network.](https://amartya18x.github.io/tapas/bnn)
For more information, refer to [our paper on arxiv](https://arxiv.org/abs/1806.03461).