Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Alvarez-Gonzalez, Nurudin, Kaltenbrunner, Andreas, Gómez, Vicenç
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2312.05905
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909187057909760
author Alvarez-Gonzalez, Nurudin
Kaltenbrunner, Andreas
Gómez, Vicenç
author_facet Alvarez-Gonzalez, Nurudin
Kaltenbrunner, Andreas
Gómez, Vicenç
contents We present a novel edge-level ego-network encoding for learning on graphs that can boost Message Passing Graph Neural Networks (MP-GNNs) by providing additional node and edge features or extending message-passing formats. The proposed encoding is sufficient to distinguish Strongly Regular Graphs, a family of challenging 3-WL equivalent graphs. We show theoretically that such encoding is more expressive than node-based sub-graph MP-GNNs. In an empirical evaluation on four benchmarks with 10 graph datasets, our results match or improve previous baselines on expressivity, graph classification, graph regression, and proximity tasks -- while reducing memory usage by 18.1x in certain real-world settings.
format Preprint
id arxiv_https___arxiv_org_abs_2312_05905
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improving Subgraph-GNNs via Edge-Level Ego-Network Encodings
Alvarez-Gonzalez, Nurudin
Kaltenbrunner, Andreas
Gómez, Vicenç
Machine Learning
Artificial Intelligence
We present a novel edge-level ego-network encoding for learning on graphs that can boost Message Passing Graph Neural Networks (MP-GNNs) by providing additional node and edge features or extending message-passing formats. The proposed encoding is sufficient to distinguish Strongly Regular Graphs, a family of challenging 3-WL equivalent graphs. We show theoretically that such encoding is more expressive than node-based sub-graph MP-GNNs. In an empirical evaluation on four benchmarks with 10 graph datasets, our results match or improve previous baselines on expressivity, graph classification, graph regression, and proximity tasks -- while reducing memory usage by 18.1x in certain real-world settings.
title Improving Subgraph-GNNs via Edge-Level Ego-Network Encodings
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2312.05905