Enregistré dans:
Détails bibliographiques
Auteurs principaux: Piccialli, Veronica, Morales, Dolores Romero, Salvatore, Cecilia
Format: Preprint
Publié: 2022
Sujets:
Accès en ligne:https://arxiv.org/abs/2211.09894
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914642427641856
author Piccialli, Veronica
Morales, Dolores Romero
Salvatore, Cecilia
author_facet Piccialli, Veronica
Morales, Dolores Romero
Salvatore, Cecilia
contents Counterfactual Explanations are becoming a de-facto standard in post-hoc interpretable machine learning. For a given classifier and an instance classified in an undesired class, its counterfactual explanation corresponds to small perturbations of that instance that allows changing the classification outcome. This work aims to leverage Counterfactual Explanations to detect the important decision boundaries of a pre-trained black-box model. This information is used to build a supervised discretization of the features in the dataset with a tunable granularity. Using the discretized dataset, an optimal Decision Tree can be trained that resembles the black-box model, but that is interpretable and compact. Numerical results on real-world datasets show the effectiveness of the approach in terms of accuracy and sparsity.
format Preprint
id arxiv_https___arxiv_org_abs_2211_09894
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Supervised Feature Compression based on Counterfactual Analysis
Piccialli, Veronica
Morales, Dolores Romero
Salvatore, Cecilia
Machine Learning
Optimization and Control
Counterfactual Explanations are becoming a de-facto standard in post-hoc interpretable machine learning. For a given classifier and an instance classified in an undesired class, its counterfactual explanation corresponds to small perturbations of that instance that allows changing the classification outcome. This work aims to leverage Counterfactual Explanations to detect the important decision boundaries of a pre-trained black-box model. This information is used to build a supervised discretization of the features in the dataset with a tunable granularity. Using the discretized dataset, an optimal Decision Tree can be trained that resembles the black-box model, but that is interpretable and compact. Numerical results on real-world datasets show the effectiveness of the approach in terms of accuracy and sparsity.
title Supervised Feature Compression based on Counterfactual Analysis
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2211.09894