Enregistré dans:
Détails bibliographiques
Auteurs principaux: Rida, Adam, Lesot, Marie-Jeanne, Renard, Xavier, Marsala, Christophe
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2309.17095
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908288847708160
author Rida, Adam
Lesot, Marie-Jeanne
Renard, Xavier
Marsala, Christophe
author_facet Rida, Adam
Lesot, Marie-Jeanne
Renard, Xavier
Marsala, Christophe
contents Explainable AI (XAI) methods have mostly been built to investigate and shed light on single machine learning models and are not designed to capture and explain differences between multiple models effectively. This paper addresses the challenge of understanding and explaining differences between machine learning models, which is crucial for model selection, monitoring and lifecycle management in real-world applications. We propose DeltaXplainer, a model-agnostic method for generating rule-based explanations describing the differences between two binary classifiers. To assess the effectiveness of DeltaXplainer, we conduct experiments on synthetic and real-world datasets, covering various model comparison scenarios involving different types of concept drift.
format Preprint
id arxiv_https___arxiv_org_abs_2309_17095
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Dynamic Interpretability for Model Comparison via Decision Rules
Rida, Adam
Lesot, Marie-Jeanne
Renard, Xavier
Marsala, Christophe
Machine Learning
Artificial Intelligence
Explainable AI (XAI) methods have mostly been built to investigate and shed light on single machine learning models and are not designed to capture and explain differences between multiple models effectively. This paper addresses the challenge of understanding and explaining differences between machine learning models, which is crucial for model selection, monitoring and lifecycle management in real-world applications. We propose DeltaXplainer, a model-agnostic method for generating rule-based explanations describing the differences between two binary classifiers. To assess the effectiveness of DeltaXplainer, we conduct experiments on synthetic and real-world datasets, covering various model comparison scenarios involving different types of concept drift.
title Dynamic Interpretability for Model Comparison via Decision Rules
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2309.17095