LIBRE: Learning Interpretable Boolean Rule Ensembles

Abstract

We present a novel method—LIBRE—learn an interpretable classifier, which materializes as a set of Boolean rules. LIBRE uses an ensemble of bottom-up, weak learners operating on a random subset of features, which allows for the learning of rules that generalize well on unseen data even in imbalanced settings. Weak learners are combined with a simple union so that the final ensemble is also interpretable. Experimental results indicate that LIBRE efficiently strikes the right balance between prediction accuracy, which is competitive with black-box methods, and interpretability, which is often superior to alternative methods from the literature.

Cite

Text

Mita et al. "LIBRE: Learning Interpretable Boolean Rule Ensembles." Artificial Intelligence and Statistics, 2020.

Markdown

[Mita et al. "LIBRE: Learning Interpretable Boolean Rule Ensembles." Artificial Intelligence and Statistics, 2020.](https://mlanthology.org/aistats/2020/mita2020aistats-libre/)

BibTeX

@inproceedings{mita2020aistats-libre,
  title     = {{LIBRE: Learning Interpretable Boolean Rule Ensembles}},
  author    = {Mita, Graziano and Papotti, Paolo and Filippone, Maurizio and Michiardi, Pietro},
  booktitle = {Artificial Intelligence and Statistics},
  year      = {2020},
  pages     = {245-255},
  volume    = {108},
  url       = {https://mlanthology.org/aistats/2020/mita2020aistats-libre/}
}