Demos and Tutorials

This section lists several demos that apply Concrete-ML to some popular machine learning problems. They show how to build ML models that perform well under FHE constraints, and then how to perform the conversion to FHE.

Simpler tutorials that discuss only model usage and compilation are also available for the built-in models and for deep learning.

Titanic

Train an XGB classifier that can perform encrypted prediction for the Kaggle Titanic competition

Neural Network Fine-tuning

Fine-tune a VGG network to classify the CIFAR image data-sets and predict on encrypted data

Neural Network Splitting for SaaS deployment

Train a VGG-like CNN that classifies CIFAR10 encrypted images, and where an initial feature extractor is executed client-side

Handwritten digit classification

Train a neural network model to classify encrypted digit images from the MNIST data-set

Encrypted Image filtering

A Hugging Face space that applies a variety of image filters to encrypted images

Encrypted sentiment analysis

A Hugging Face space that securely analyzes the sentiment expressed in a short text