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 e↵ects of medications. In this contribution, we develop a POMDPbased 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.
Document type :
Conference papers
Complete list of metadatas
Contributor : Yann Chevaleyre <>
Submitted on : Monday, February 4, 2019 - 3:02:23 PM
Last modification on : Wednesday, February 19, 2020 - 8:58:36 AM


  • HAL Id : hal-02006252, version 1


Matthieu Clertant, Nataliya Sokolovska, Yann Chevaleyre, Blaise Hanczar. Interpretable Cascade Classifiers with Abstention. 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019), Apr 2019, Naha, Japan. ⟨hal-02006252⟩



Record views