Enhancing Locally Adaptive Smoothing of Graph Neural Networks Via Laplacian Node Disagreement

Wang, Y; Hu, L; Cao, X; Chang, Y; Tsang, IW

Enhancing Locally Adaptive Smoothing of Graph Neural Networks Via Laplacian Node Disagreement

Wang, Y Hu, L Cao, X Chang, Y Tsang, IW

Permalink

Publisher:: IEEE COMPUTER SOC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Knowledge and Data Engineering, 2024, 36, (3), pp. 1099-1112
Issue Date:: 2024-03-01

Recently Added

	Filename	Description	Size
	1662105.pdf	Published version	2.33 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is new to OPUS and is not currently available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wang, Y
dc.contributor.author	Hu, L
dc.contributor.author	Cao, X
dc.contributor.author	Chang, Y
dc.contributor.author	Tsang, IW
dc.date.accessioned	2024-08-07T04:23:24Z
dc.date.available	2024-08-07T04:23:24Z
dc.date.issued	2024-03-01
dc.identifier.citation	IEEE Transactions on Knowledge and Data Engineering, 2024, 36, (3), pp. 1099-1112
dc.identifier.issn	1041-4347
dc.identifier.issn	1558-2191
dc.identifier.uri	http://hdl.handle.net/10453/180181
dc.description.abstract	Graph neural networks (GNNs) are designed to perform inference on data described by graph-structured node features and topology information. From the perspective of graph signal denoising, the typical message passing schemes of GNNs act as a globally uniform smoothing that minimizes disagreements between embeddings of connected nodes. However, the level of smoothing over different regions of the graph should be different, especially for those inter-class regions. This deviation limits the expressiveness of GNNs, and then renders them fragile to over-smoothing, long-range dependencies, and non-homophily settings. In this paper, we find that the node disagreements of initial graph features can present more trustworthy constraints on node embeddings, thereby enhancing the locally adaptive smoothing of GNNs. To spread the inherent disagreements of nodes, we propose the Laplacian node disagreement to jointly measure the initial features and output embeddings. With such a measurement, we then present a new graph signal denoising objective deriving a more effective message passing scheme and further incorporate it into the GNN architecture, named Laplacian node disagreement-based GNN (LND-GNN). Learning from its output node representations, we integrate an auxiliary disagreement constraint into the overall classification loss. Experiments demonstrate the expressive ability of LND-GNN in the downstream semi-supervised node classification task.
dc.language	English
dc.publisher	IEEE COMPUTER SOC
dc.relation.ispartof	IEEE Transactions on Knowledge and Data Engineering
dc.relation.isbasedon	10.1109/TKDE.2023.3303212
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.subject	08 Information and Computing Sciences
dc.subject.classification	Information Systems
dc.subject.classification	46 Information and computing sciences
dc.title	Enhancing Locally Adaptive Smoothing of Graph Neural Networks Via Laplacian Node Disagreement
dc.type	Journal Article
utslib.citation.volume	36
utslib.for	08 Information and Computing Sciences
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Strength - AAII - Australian Artificial Intelligence Institute
pubs.organisational-group	University of Technology Sydney/All Manual Groups
pubs.organisational-group	University of Technology Sydney/All Manual Groups/Australian Artificial Intelligence Institute (AAII)
utslib.copyright.status	recently_added	*
dc.date.updated	2024-08-07T04:23:21Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	36
utslib.citation.issue	3

Abstract:

Graph neural networks (GNNs) are designed to perform inference on data described by graph-structured node features and topology information. From the perspective of graph signal denoising, the typical message passing schemes of GNNs act as a globally uniform smoothing that minimizes disagreements between embeddings of connected nodes. However, the level of smoothing over different regions of the graph should be different, especially for those inter-class regions. This deviation limits the expressiveness of GNNs, and then renders them fragile to over-smoothing, long-range dependencies, and non-homophily settings. In this paper, we find that the node disagreements of initial graph features can present more trustworthy constraints on node embeddings, thereby enhancing the locally adaptive smoothing of GNNs. To spread the inherent disagreements of nodes, we propose the Laplacian node disagreement to jointly measure the initial features and output embeddings. With such a measurement, we then present a new graph signal denoising objective deriving a more effective message passing scheme and further incorporate it into the GNN architecture, named Laplacian node disagreement-based GNN (LND-GNN). Learning from its output node representations, we integrate an auxiliary disagreement constraint into the overall classification loss. Experiments demonstrate the expressive ability of LND-GNN in the downstream semi-supervised node classification task.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/180181