This notebook introduces a Privacy-Preserving Machine Learning (PPML) solution to the Kaggle Titanic competition using the Concrete ML open-source framework. Its main ambition is to show that Fully Homomorphic Encryption (FHE) can be used for protecting data when using a Machine Learning model to predict outcomes without degrading its performance. In this example, a XGBoost classifier model will be considered as it achieves near state-of-the-art accuracy.
- First, create a virtual env and activate it:
python3.8 -m venv .venv
source .venv/bin/activate- Then, install required packages:
pip3 install -U pip wheel setuptools --ignore-installed
pip3 install -r requirements.txt --ignore-installed