Interpretable Cascade Classifiers with Abstention
Abstract
In many prediction tasks such as medical diagnostics, sequential decisions are crucial to provide optimal individual treatment. Budget in real-life applications is always limited, and it can represent any limited resource such as time, money, or side effects of medications. In this contribution, we develop a POMDP-based framework to learn cost-sensitive heterogeneous cascading systems. We provide both the theoretical support for the introduced approach and the intuition behind it. We evaluate our novel method on some standard benchmarks, and we discuss how the learned models can be interpreted by human experts.
Cite
Text
Clertant et al. "Interpretable Cascade Classifiers with Abstention." Artificial Intelligence and Statistics, 2019.Markdown
[Clertant et al. "Interpretable Cascade Classifiers with Abstention." Artificial Intelligence and Statistics, 2019.](https://mlanthology.org/aistats/2019/clertant2019aistats-interpretable/)BibTeX
@inproceedings{clertant2019aistats-interpretable,
title = {{Interpretable Cascade Classifiers with Abstention}},
author = {Clertant, Matthieu and Sokolovska, Nataliya and Chevaleyre, Yann and Hanczar, Blaise},
booktitle = {Artificial Intelligence and Statistics},
year = {2019},
pages = {2312-2320},
volume = {89},
url = {https://mlanthology.org/aistats/2019/clertant2019aistats-interpretable/}
}