https://github.com/alessandro-nori/crypto_nn

Proof of concept for CryptoDL made for BigSec course @ EURECOM
https://github.com/alessandro-nori/crypto_nn

homomorphic-encryption machine-learning neural-network

Last synced: 4 months ago
JSON representation

Proof of concept for CryptoDL made for BigSec course @ EURECOM

• Host: GitHub
• URL: https://github.com/alessandro-nori/crypto_nn
• Owner: alessandro-nori
• Created: 2019-01-12T15:22:58.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2019-05-14T14:56:33.000Z (about 7 years ago)
• Last Synced: 2024-07-31T16:03:17.562Z (almost 2 years ago)
• Topics: homomorphic-encryption, machine-learning, neural-network
• Language: C++
• Homepage:
• Size: 33.2 KB
• Stars: 18
• Watchers: 1
• Forks: 3
• Open Issues: 2
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Crypto Neural Network

Crypto_nn is a very simple example of neural network that can perform classification over encrypted data using homomorphic encryption.

The idea is taken from [CryptoDL: Deep Neural Networks over Encrypted Data](https://arxiv.org/pdf/1711.05189.pdf) by Ehsan Hesamifard, Hassan Takabi, Mehdi Ghasemi where

you can find all the details.

## Activation function

To use activation functions within HE schemes, they should be approximated in a form which is implemented using only addition and multiplication (e.g. polynomial).

In this example I simulated the ReLU function as presented in the paper and I obtained the following approximation:

0.0012x2 + 0.5x + 52

![alt text](https://github.com/ironordnassela/crypto_nn/blob/master/relu/relu_graphic.png)