Saved in:
Bibliographic Details
Main Authors: Liu, Mingyue, Cropper, Andrew
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.24633
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915583259312128
author Liu, Mingyue
Cropper, Andrew
author_facet Liu, Mingyue
Cropper, Andrew
contents Inductive logic programming (ILP) is a form of logical machine learning. Most ILP algorithms learn a single hypothesis from a single training run. Ensemble methods train an ILP algorithm multiple times to learn multiple hypotheses. In this paper, we train an ILP algorithm only once and save intermediate hypotheses. We then combine the hypotheses using a minimum description length weighting scheme. Our experiments on multiple benchmarks, including game playing and visual reasoning, show that our approach improves predictive accuracy by 4% with less than 1% computational overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Symbolic Snapshot Ensembles
Liu, Mingyue
Cropper, Andrew
Machine Learning
Logic in Computer Science
Inductive logic programming (ILP) is a form of logical machine learning. Most ILP algorithms learn a single hypothesis from a single training run. Ensemble methods train an ILP algorithm multiple times to learn multiple hypotheses. In this paper, we train an ILP algorithm only once and save intermediate hypotheses. We then combine the hypotheses using a minimum description length weighting scheme. Our experiments on multiple benchmarks, including game playing and visual reasoning, show that our approach improves predictive accuracy by 4% with less than 1% computational overhead.
title Symbolic Snapshot Ensembles
topic Machine Learning
Logic in Computer Science
url https://arxiv.org/abs/2510.24633