We present an approach to improve the accuracy-interpretability trade-off of Machine Learning (ML) Decision Trees (DTs). In particular, we apply Maximum Satisfiability technology to compute Minimum Pure DTs (MPDTs). We improve the runtime of previous approaches and, show that these MPDTs can outperform the accuracy of DTs generated with the ML framework sklearn.
@article{arxiv.2110.13854,
title = {Interpretable Decision Trees Through MaxSAT},
author = {Josep Alos and Carlos Ansotegui and Eduard Torres},
journal= {arXiv preprint arXiv:2110.13854},
year = {2022}
}