Counterfactual Metarules for Local and Global Recourse

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bewley, Tom, Amoukou, Salim I., Mishra, Saumitra, Magazzeni, Daniele, Veloso, Manuela
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911892166934528
author Bewley, Tom
Amoukou, Salim I.
Mishra, Saumitra
Magazzeni, Daniele
Veloso, Manuela
author_facet Bewley, Tom
Amoukou, Salim I.
Mishra, Saumitra
Magazzeni, Daniele
Veloso, Manuela
contents We introduce T-CREx, a novel model-agnostic method for local and global counterfactual explanation (CE), which summarises recourse options for both individuals and groups in the form of human-readable rules. It leverages tree-based surrogate models to learn the counterfactual rules, alongside 'metarules' denoting their regions of optimality, providing both a global analysis of model behaviour and diverse recourse options for users. Experiments indicate that T-CREx achieves superior aggregate performance over existing rule-based baselines on a range of CE desiderata, while being orders of magnitude faster to run.
format Preprint
id arxiv_https___arxiv_org_abs_2405_18875
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Counterfactual Metarules for Local and Global Recourse
Bewley, Tom
Amoukou, Salim I.
Mishra, Saumitra
Magazzeni, Daniele
Veloso, Manuela
Artificial Intelligence
We introduce T-CREx, a novel model-agnostic method for local and global counterfactual explanation (CE), which summarises recourse options for both individuals and groups in the form of human-readable rules. It leverages tree-based surrogate models to learn the counterfactual rules, alongside 'metarules' denoting their regions of optimality, providing both a global analysis of model behaviour and diverse recourse options for users. Experiments indicate that T-CREx achieves superior aggregate performance over existing rule-based baselines on a range of CE desiderata, while being orders of magnitude faster to run.
title Counterfactual Metarules for Local and Global Recourse
topic Artificial Intelligence
url https://arxiv.org/abs/2405.18875