Relating-Up: Advancing Graph Neural Networks through Inter-Graph Relationships

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zou, Qi, Yu, Na, Zhang, Daoliang, Zhang, Wei, Gao, Rui
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913343253512192
author Zou, Qi
Yu, Na
Zhang, Daoliang
Zhang, Wei
Gao, Rui
author_facet Zou, Qi
Yu, Na
Zhang, Daoliang
Zhang, Wei
Gao, Rui
contents Graph Neural Networks (GNNs) have excelled in learning from graph-structured data, especially in understanding the relationships within a single graph, i.e., intra-graph relationships. Despite their successes, GNNs are limited by neglecting the context of relationships across graphs, i.e., inter-graph relationships. Recognizing the potential to extend this capability, we introduce Relating-Up, a plug-and-play module that enhances GNNs by exploiting inter-graph relationships. This module incorporates a relation-aware encoder and a feedback training strategy. The former enables GNNs to capture relationships across graphs, enriching relation-aware graph representation through collective context. The latter utilizes a feedback loop mechanism for the recursively refinement of these representations, leveraging insights from refining inter-graph dynamics to conduct feedback loop. The synergy between these two innovations results in a robust and versatile module. Relating-Up enhances the expressiveness of GNNs, enabling them to encapsulate a wider spectrum of graph relationships with greater precision. Our evaluations across 16 benchmark datasets demonstrate that integrating Relating-Up into GNN architectures substantially improves performance, positioning Relating-Up as a formidable choice for a broad spectrum of graph representation learning tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03950
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Relating-Up: Advancing Graph Neural Networks through Inter-Graph Relationships
Zou, Qi
Yu, Na
Zhang, Daoliang
Zhang, Wei
Gao, Rui
Machine Learning
Artificial Intelligence
Graph Neural Networks (GNNs) have excelled in learning from graph-structured data, especially in understanding the relationships within a single graph, i.e., intra-graph relationships. Despite their successes, GNNs are limited by neglecting the context of relationships across graphs, i.e., inter-graph relationships. Recognizing the potential to extend this capability, we introduce Relating-Up, a plug-and-play module that enhances GNNs by exploiting inter-graph relationships. This module incorporates a relation-aware encoder and a feedback training strategy. The former enables GNNs to capture relationships across graphs, enriching relation-aware graph representation through collective context. The latter utilizes a feedback loop mechanism for the recursively refinement of these representations, leveraging insights from refining inter-graph dynamics to conduct feedback loop. The synergy between these two innovations results in a robust and versatile module. Relating-Up enhances the expressiveness of GNNs, enabling them to encapsulate a wider spectrum of graph relationships with greater precision. Our evaluations across 16 benchmark datasets demonstrate that integrating Relating-Up into GNN architectures substantially improves performance, positioning Relating-Up as a formidable choice for a broad spectrum of graph representation learning tasks.
title Relating-Up: Advancing Graph Neural Networks through Inter-Graph Relationships
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2405.03950