Symmetry breaking for inductive logic programming

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cropper, Andrew, Cerna, David M., Järvisalo, Matti
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911392826654720
author Cropper, Andrew
Cerna, David M.
Järvisalo, Matti
author_facet Cropper, Andrew
Cerna, David M.
Järvisalo, Matti
contents The goal of inductive logic programming is to search for a hypothesis that generalises training data and background knowledge. The challenge is searching vast hypothesis spaces, which is exacerbated because many logically equivalent hypotheses exist. To address this challenge, we introduce a method to break symmetries in the hypothesis space. We implement our idea in answer set programming. Our experiments on multiple domains, including visual reasoning and game playing, show that our approach can reduce solving times from over an hour to just 17 seconds.
format Preprint
id arxiv_https___arxiv_org_abs_2508_06263
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Symmetry breaking for inductive logic programming
Cropper, Andrew
Cerna, David M.
Järvisalo, Matti
Artificial Intelligence
Machine Learning
The goal of inductive logic programming is to search for a hypothesis that generalises training data and background knowledge. The challenge is searching vast hypothesis spaces, which is exacerbated because many logically equivalent hypotheses exist. To address this challenge, we introduce a method to break symmetries in the hypothesis space. We implement our idea in answer set programming. Our experiments on multiple domains, including visual reasoning and game playing, show that our approach can reduce solving times from over an hour to just 17 seconds.
title Symmetry breaking for inductive logic programming
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2508.06263