Protected probabilistic classification experiments

This repository contains the supporting code for numerical experiments in the Chapter "Protected Probabilistic Classification Library" in the book "The Importance of Being Learnable" by Springer 2026.

The experiments were carried out using Python 3.11.8 and make use of the protected-classification library (https://github.com/ip200/protected-classification)

Instructions

Please install the necessary requirements:

pip install -r requirements.txt

In order to run the experiments for a given section please run

python section_5_1_1.py

The results will be stored in the ./results folder with the corresponding tex formatted output in the ./results/experiment/tex folder (e.g ./results/section_5_1_1/tex)

Please note that experiments for Section 5.1.2 are not possible to be shared here due to the large dataset size > 25 MB