Sound Logical Explanations for Mean Aggregation Graph Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Morris, Matthew, Horrocks, Ian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912710061457408
author Morris, Matthew
Horrocks, Ian
author_facet Morris, Matthew
Horrocks, Ian
contents Graph neural networks (GNNs) are frequently used for knowledge graph completion. Their black-box nature has motivated work that uses sound logical rules to explain predictions and characterise their expressivity. However, despite the prevalence of GNNs that use mean as an aggregation function, explainability and expressivity results are lacking for them. We consider GNNs with mean aggregation and non-negative weights (MAGNNs), proving the precise class of monotonic rules that can be sound for them, as well as providing a restricted fragment of first-order logic to explain any MAGNN prediction. Our experiments show that restricting mean-aggregation GNNs to have non-negative weights yields comparable or improved performance on standard inductive benchmarks, that sound rules are obtained in practice, that insightful explanations can be generated in practice, and that the sound rules can expose issues in the trained models.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11593
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sound Logical Explanations for Mean Aggregation Graph Neural Networks
Morris, Matthew
Horrocks, Ian
Machine Learning
Artificial Intelligence
Logic in Computer Science
03B70
I.2.6; G.2.2; I.2.4; I.2.3
Graph neural networks (GNNs) are frequently used for knowledge graph completion. Their black-box nature has motivated work that uses sound logical rules to explain predictions and characterise their expressivity. However, despite the prevalence of GNNs that use mean as an aggregation function, explainability and expressivity results are lacking for them. We consider GNNs with mean aggregation and non-negative weights (MAGNNs), proving the precise class of monotonic rules that can be sound for them, as well as providing a restricted fragment of first-order logic to explain any MAGNN prediction. Our experiments show that restricting mean-aggregation GNNs to have non-negative weights yields comparable or improved performance on standard inductive benchmarks, that sound rules are obtained in practice, that insightful explanations can be generated in practice, and that the sound rules can expose issues in the trained models.
title Sound Logical Explanations for Mean Aggregation Graph Neural Networks
topic Machine Learning
Artificial Intelligence
Logic in Computer Science
03B70
I.2.6; G.2.2; I.2.4; I.2.3
url https://arxiv.org/abs/2511.11593