SkyMap: a generative graph model for GNN benchmarking

Graph Neural Networks (GNNs) have gained considerable attention in recent years. Despite the surge in innovative GNN architecture designs, research heavily relies on the same 5-10 benchmark datasets for validation. To address this limitation, several generative graph models like ALBTER or GenCAT hav...

Descripción completa

Detalles Bibliográficos
Autores: Wassington, Axel, Abadal Cavallé, Sergi|||0000-0003-0941-0260, Higueras, Raúl
Tipo de recurso: artículo
Fecha de publicación:2024
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/419897
Acceso en línea:https://hdl.handle.net/2117/419897
https://dx.doi.org/10.3389/frai.2024.1427534
Access Level:acceso abierto
Palabra clave:Graph Neural Network (GNN)
Machine learning datasets
Graph generation model
Mixing matrix
Degree distribution
Benchmark
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
id ES_be4e8890a2a03a5589d6df194e9501d8
oai_identifier_str oai:upcommons.upc.edu:2117/419897
network_acronym_str ES
network_name_str España
repository_id_str
spelling SkyMap: a generative graph model for GNN benchmarkingWassington, AxelAbadal Cavallé, Sergi|||0000-0003-0941-0260Higueras, RaúlGraph Neural Network (GNN)Machine learning datasetsGraph generation modelMixing matrixDegree distributionBenchmarkÀrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàticGraph Neural Networks (GNNs) have gained considerable attention in recent years. Despite the surge in innovative GNN architecture designs, research heavily relies on the same 5-10 benchmark datasets for validation. To address this limitation, several generative graph models like ALBTER or GenCAT have emerged, aiming to fix this problem with synthetic graph datasets. However, these models often struggle to mirror the GNN performance of the original graphs. In this work, we present SkyMap, a generative model for labeled attributed graphs with a fine-grained control over graph topology and feature distribution parameters. We show that our model is able to consistently replicate the learnability of graphs on graph convolutional, attention, and isomorphism networks better (64% lower Wasserstein distance) than ALBTER and GenCAT. Further, we prove that by randomly sampling the input parameters of SkyMap, graph dataset constellations can be created that cover a large parametric space, hence making a significant stride in crafting synthetic datasets tailored for GNN evaluation and benchmarking, as we illustrate through a performance comparison between a GNN and a multilayer perceptron.Peer ReviewedFrontiers Media SA20242024-11-1420242024-12-05journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/2117/419897https://dx.doi.org/10.3389/frai.2024.1427534reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2Attribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/4198972026-05-27T15:37:01Z
dc.title.none.fl_str_mv SkyMap: a generative graph model for GNN benchmarking
title SkyMap: a generative graph model for GNN benchmarking
spellingShingle SkyMap: a generative graph model for GNN benchmarking
Wassington, Axel
Graph Neural Network (GNN)
Machine learning datasets
Graph generation model
Mixing matrix
Degree distribution
Benchmark
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
title_short SkyMap: a generative graph model for GNN benchmarking
title_full SkyMap: a generative graph model for GNN benchmarking
title_fullStr SkyMap: a generative graph model for GNN benchmarking
title_full_unstemmed SkyMap: a generative graph model for GNN benchmarking
title_sort SkyMap: a generative graph model for GNN benchmarking
dc.creator.none.fl_str_mv Wassington, Axel
Abadal Cavallé, Sergi|||0000-0003-0941-0260
Higueras, Raúl
author Wassington, Axel
author_facet Wassington, Axel
Abadal Cavallé, Sergi|||0000-0003-0941-0260
Higueras, Raúl
author_role author
author2 Abadal Cavallé, Sergi|||0000-0003-0941-0260
Higueras, Raúl
author2_role author
author
dc.subject.none.fl_str_mv Graph Neural Network (GNN)
Machine learning datasets
Graph generation model
Mixing matrix
Degree distribution
Benchmark
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
topic Graph Neural Network (GNN)
Machine learning datasets
Graph generation model
Mixing matrix
Degree distribution
Benchmark
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
description Graph Neural Networks (GNNs) have gained considerable attention in recent years. Despite the surge in innovative GNN architecture designs, research heavily relies on the same 5-10 benchmark datasets for validation. To address this limitation, several generative graph models like ALBTER or GenCAT have emerged, aiming to fix this problem with synthetic graph datasets. However, these models often struggle to mirror the GNN performance of the original graphs. In this work, we present SkyMap, a generative model for labeled attributed graphs with a fine-grained control over graph topology and feature distribution parameters. We show that our model is able to consistently replicate the learnability of graphs on graph convolutional, attention, and isomorphism networks better (64% lower Wasserstein distance) than ALBTER and GenCAT. Further, we prove that by randomly sampling the input parameters of SkyMap, graph dataset constellations can be created that cover a large parametric space, hence making a significant stride in crafting synthetic datasets tailored for GNN evaluation and benchmarking, as we illustrate through a performance comparison between a GNN and a multilayer perceptron.
publishDate 2024
dc.date.none.fl_str_mv 2024
2024-11-14
2024
2024-12-05
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/419897
https://dx.doi.org/10.3389/frai.2024.1427534
url https://hdl.handle.net/2117/419897
https://dx.doi.org/10.3389/frai.2024.1427534
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Frontiers Media SA
publisher.none.fl_str_mv Frontiers Media SA
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418275377512448
score 15,812429

Similares en LA Referencia