Why this and not that? A Logic-based Framework for Contrastive Explanations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Geibinger, Tobias, Jaakkola, Reijo, Kuusisto, Antti, Liu, Xinghan, Vilander, Miikka
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912476394684416
author Geibinger, Tobias
Jaakkola, Reijo
Kuusisto, Antti
Liu, Xinghan
Vilander, Miikka
author_facet Geibinger, Tobias
Jaakkola, Reijo
Kuusisto, Antti
Liu, Xinghan
Vilander, Miikka
contents We define several canonical problems related to contrastive explanations, each answering a question of the form ''Why P but not Q?''. The problems compute causes for both P and Q, explicitly comparing their differences. We investigate the basic properties of our definitions in the setting of propositional logic. We show, inter alia, that our framework captures a cardinality-minimal version of existing contrastive explanations in the literature. Furthermore, we provide an extensive analysis of the computational complexities of the problems. We also implement the problems for CNF-formulas using answer set programming and present several examples demonstrating how they work in practice.
format Preprint
id arxiv_https___arxiv_org_abs_2507_08454
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Why this and not that? A Logic-based Framework for Contrastive Explanations
Geibinger, Tobias
Jaakkola, Reijo
Kuusisto, Antti
Liu, Xinghan
Vilander, Miikka
Artificial Intelligence
Machine Learning
Logic in Computer Science
68T27, 03B05
I.2.3; F.4.1
We define several canonical problems related to contrastive explanations, each answering a question of the form ''Why P but not Q?''. The problems compute causes for both P and Q, explicitly comparing their differences. We investigate the basic properties of our definitions in the setting of propositional logic. We show, inter alia, that our framework captures a cardinality-minimal version of existing contrastive explanations in the literature. Furthermore, we provide an extensive analysis of the computational complexities of the problems. We also implement the problems for CNF-formulas using answer set programming and present several examples demonstrating how they work in practice.
title Why this and not that? A Logic-based Framework for Contrastive Explanations
topic Artificial Intelligence
Machine Learning
Logic in Computer Science
68T27, 03B05
I.2.3; F.4.1
url https://arxiv.org/abs/2507.08454