Graph representations for reinforcement learning

Autores
Schab, Esteban Alejandro; Casanova Pietroboni, Carlos Antonio; Piccoli, María Fabiana
Año de publicación
2024
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Graph analysis is becoming increasingly important due to the expressive power of graph models and the efficient algorithms available for processing them. Reinforcement Learning is one domain that could ben- efit from advancements in graph analysis, given that a learning agent may be integrated into an environ- ment that can be represented as a graph. Nevertheless, the structural irregularity of graphs and the lack of prior labels make it difficult to integrate such a model into modern Reinforcement Learning frameworks that rely on artificial neural networks. Graph embedding enables the learning of low-dimensional vector representations that are more suited for machine learning algorithms, while retaining essential graph features. This paper presents a framework for evaluating graph embedding algorithms and their ability to preserve the structure and relevant features of graphs by means of an internal validation metric, without resorting to subsequent tasks that require labels for training. Based on this framework, three defined algorithms that meet the necessary requirements for solving a specific problem of Reinforcement Learningin graphs are selected, analyzed, and compared. These algorithms are Graph2Vec, GL2Vec, and Wavelet Characteristics, with the latter two demonstrating superior performance.
Fil: Schab, Esteban Alejandro. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.
Fil: Casanova Pietroboni, Carlos Antonio. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.
Fil: Casanova Pietroboni, Carlos Antonio. Universidad Autónoma de Entre Ríos; Argentina.
Fil: Piccoli, María Fabiana. Universidad Nacional de San Luis; Argentina.
Fil: Piccoli, María Fabiana. Universidad Autónoma de Entre Ríos; Argentina.
Fuente
Journal of Computer Science and Technology, 24(1), 29-38. (2024)
Materia
Computational intelligence
Reinforce- ment learning
Graph embeddings
Unsupervised GRL
Whole graph embedding
Nivel de accesibilidad
acceso abierto
Condiciones de uso
Attribution-NonCommercial-NoDerivatives 4.0 International
Repositorio
Repositorio Institucional Abierto (UTN)
Institución
Universidad Tecnológica Nacional
OAI Identificador
oai:ria.utn.edu.ar:20.500.12272/12050
Acceder
id RIAUTN_c1825be0d94ecd51a6d8c5e77bbcc906
oai_identifier_str oai:ria.utn.edu.ar:20.500.12272/12050
network_acronym_str RIAUTN
repository_id_str a
network_name_str Repositorio Institucional Abierto (UTN)
spelling Graph representations for reinforcement learningSchab, Esteban AlejandroCasanova Pietroboni, Carlos AntonioPiccoli, María FabianaComputational intelligenceReinforce- ment learningGraph embeddingsUnsupervised GRLWhole graph embeddingGraph analysis is becoming increasingly important due to the expressive power of graph models and the efficient algorithms available for processing them. Reinforcement Learning is one domain that could ben- efit from advancements in graph analysis, given that a learning agent may be integrated into an environ- ment that can be represented as a graph. Nevertheless, the structural irregularity of graphs and the lack of prior labels make it difficult to integrate such a model into modern Reinforcement Learning frameworks that rely on artificial neural networks. Graph embedding enables the learning of low-dimensional vector representations that are more suited for machine learning algorithms, while retaining essential graph features. This paper presents a framework for evaluating graph embedding algorithms and their ability to preserve the structure and relevant features of graphs by means of an internal validation metric, without resorting to subsequent tasks that require labels for training. Based on this framework, three defined algorithms that meet the necessary requirements for solving a specific problem of Reinforcement Learningin graphs are selected, analyzed, and compared. These algorithms are Graph2Vec, GL2Vec, and Wavelet Characteristics, with the latter two demonstrating superior performance.Fil: Schab, Esteban Alejandro. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.Fil: Casanova Pietroboni, Carlos Antonio. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.Fil: Casanova Pietroboni, Carlos Antonio. Universidad Autónoma de Entre Ríos; Argentina.Fil: Piccoli, María Fabiana. Universidad Nacional de San Luis; Argentina.Fil: Piccoli, María Fabiana. Universidad Autónoma de Entre Ríos; Argentina.Universidad Nacional de La Plata. Facultad de Informática.2024-12-27T15:06:20Z2024-04info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articulopdfapplication/pdfJournal of Computer Science and Technology1666-6038http://hdl.handle.net/20.500.12272/12050https://doi.org/10.24215/16666038.24.e03Journal of Computer Science and Technology, 24(1), 29-38. (2024)reponame:Repositorio Institucional Abierto (UTN)instname:Universidad Tecnológica Nacionalenginfo:eu-repo/semantics/openAccessAttribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/Schab, Esteban Alejandro ; Casanova Pietroboni, Carlos Antonio ; Piccoli, María Fabiana.No comercial con fines académicos. Licencia Creative Commons CC BY-NC-SA.2026-09-24T12:47:02Zoai:ria.utn.edu.ar:20.500.12272/12050instacron:UTNInstitucionalhttp://ria.utn.edu.ar/Universidad públicaNo correspondehttp://ria.utn.edu.ar/oaigestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:a2026-09-24 12:47:03.464Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacionalfalse
dc.title.none.fl_str_mv Graph representations for reinforcement learning
title Graph representations for reinforcement learning
spellingShingle Graph representations for reinforcement learning
Schab, Esteban Alejandro
Computational intelligence
Reinforce- ment learning
Graph embeddings
Unsupervised GRL
Whole graph embedding
title_short Graph representations for reinforcement learning
title_full Graph representations for reinforcement learning
title_fullStr Graph representations for reinforcement learning
title_full_unstemmed Graph representations for reinforcement learning
title_sort Graph representations for reinforcement learning
dc.creator.none.fl_str_mv Schab, Esteban Alejandro
Casanova Pietroboni, Carlos Antonio
Piccoli, María Fabiana
author Schab, Esteban Alejandro
author_facet Schab, Esteban Alejandro
Casanova Pietroboni, Carlos Antonio
Piccoli, María Fabiana
author_role author
author2 Casanova Pietroboni, Carlos Antonio
Piccoli, María Fabiana
author2_role author
author
dc.subject.none.fl_str_mv Computational intelligence
Reinforce- ment learning
Graph embeddings
Unsupervised GRL
Whole graph embedding
topic Computational intelligence
Reinforce- ment learning
Graph embeddings
Unsupervised GRL
Whole graph embedding
dc.description.none.fl_txt_mv Graph analysis is becoming increasingly important due to the expressive power of graph models and the efficient algorithms available for processing them. Reinforcement Learning is one domain that could ben- efit from advancements in graph analysis, given that a learning agent may be integrated into an environ- ment that can be represented as a graph. Nevertheless, the structural irregularity of graphs and the lack of prior labels make it difficult to integrate such a model into modern Reinforcement Learning frameworks that rely on artificial neural networks. Graph embedding enables the learning of low-dimensional vector representations that are more suited for machine learning algorithms, while retaining essential graph features. This paper presents a framework for evaluating graph embedding algorithms and their ability to preserve the structure and relevant features of graphs by means of an internal validation metric, without resorting to subsequent tasks that require labels for training. Based on this framework, three defined algorithms that meet the necessary requirements for solving a specific problem of Reinforcement Learningin graphs are selected, analyzed, and compared. These algorithms are Graph2Vec, GL2Vec, and Wavelet Characteristics, with the latter two demonstrating superior performance.
Fil: Schab, Esteban Alejandro. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.
Fil: Casanova Pietroboni, Carlos Antonio. Universidad Tecnológica Nacional. Facultad Regional Concepción del Uruguay. Departamento Ingeniería en Sistemas de Información. Grupo de Investigación Inteligencia Computacional e Ingeniería de Software; Argentina.
Fil: Casanova Pietroboni, Carlos Antonio. Universidad Autónoma de Entre Ríos; Argentina.
Fil: Piccoli, María Fabiana. Universidad Nacional de San Luis; Argentina.
Fil: Piccoli, María Fabiana. Universidad Autónoma de Entre Ríos; Argentina.
description Graph analysis is becoming increasingly important due to the expressive power of graph models and the efficient algorithms available for processing them. Reinforcement Learning is one domain that could ben- efit from advancements in graph analysis, given that a learning agent may be integrated into an environ- ment that can be represented as a graph. Nevertheless, the structural irregularity of graphs and the lack of prior labels make it difficult to integrate such a model into modern Reinforcement Learning frameworks that rely on artificial neural networks. Graph embedding enables the learning of low-dimensional vector representations that are more suited for machine learning algorithms, while retaining essential graph features. This paper presents a framework for evaluating graph embedding algorithms and their ability to preserve the structure and relevant features of graphs by means of an internal validation metric, without resorting to subsequent tasks that require labels for training. Based on this framework, three defined algorithms that meet the necessary requirements for solving a specific problem of Reinforcement Learningin graphs are selected, analyzed, and compared. These algorithms are Graph2Vec, GL2Vec, and Wavelet Characteristics, with the latter two demonstrating superior performance.
publishDate 2024
dc.date.none.fl_str_mv 2024-12-27T15:06:20Z
2024-04
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv Journal of Computer Science and Technology
1666-6038
http://hdl.handle.net/20.500.12272/12050
https://doi.org/10.24215/16666038.24.e03
identifier_str_mv Journal of Computer Science and Technology
1666-6038
url http://hdl.handle.net/20.500.12272/12050
https://doi.org/10.24215/16666038.24.e03
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
Schab, Esteban Alejandro ; Casanova Pietroboni, Carlos Antonio ; Piccoli, María Fabiana.
No comercial con fines académicos. Licencia Creative Commons CC BY-NC-SA.
eu_rights_str_mv openAccess
rights_invalid_str_mv Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
Schab, Esteban Alejandro ; Casanova Pietroboni, Carlos Antonio ; Piccoli, María Fabiana.
No comercial con fines académicos. Licencia Creative Commons CC BY-NC-SA.
dc.format.none.fl_str_mv pdf
application/pdf
dc.publisher.none.fl_str_mv Universidad Nacional de La Plata. Facultad de Informática.
publisher.none.fl_str_mv Universidad Nacional de La Plata. Facultad de Informática.
dc.source.none.fl_str_mv Journal of Computer Science and Technology, 24(1), 29-38. (2024)
reponame:Repositorio Institucional Abierto (UTN)
instname:Universidad Tecnológica Nacional
reponame_str Repositorio Institucional Abierto (UTN)
collection Repositorio Institucional Abierto (UTN)
instname_str Universidad Tecnológica Nacional
repository.name.fl_str_mv Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacional
repository.mail.fl_str_mv gestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.ar
_version_ 1877230932243513344
score 13.265058

Publicaciones similares