Energy-Efficient Federated Learning for AIoT Using Clustering Methods

While substantial research has been devoted to optimizing model performance, convergence rates, and communication efficiency, the energy implications of federated learning (FL) within Artificial Intelligence of Things (AIoT) scenarios are often overlooked in the existing literature. This study exami...

Descripción completa

Detalles Bibliográficos
Autores: Pereira R., Fama F., Kalalas C., Dini P.
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2025
País:España
Institución:Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
Repositorio:r-CTTC. Repositorio Institucional Producción Científica del Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
OAI Identifier:oai:cttc.fundanetsuite.com:p8699
Acceso en línea:https://cttc.fundanetsuite.com/Publicaciones/ProdCientif/PublicacionFrw.aspx?id=8699
Access Level:acceso abierto
Palabra clave:Training
Internet of Things
Artificial intelligence
Convergence
Data models
Servers
Computational modeling
Costs
Distributed databases
Energy consumption
Artificial Intelligence of Things (AIoT) device selection
clustering
energy efficiency
federated learning (FL)
id ES_abdc726cd78ef37cc1ccf97e1498a239
oai_identifier_str oai:cttc.fundanetsuite.com:p8699
network_acronym_str ES
network_name_str España
repository_id_str
spelling Energy-Efficient Federated Learning for AIoT Using Clustering MethodsPereira R.Fama F.Kalalas C.Dini P.TrainingInternet of ThingsArtificial intelligenceConvergenceData modelsServersComputational modelingCostsDistributed databasesEnergy consumptionArtificial Intelligence of Things (AIoT) device selectionclusteringenergy efficiencyfederated learning (FL)While substantial research has been devoted to optimizing model performance, convergence rates, and communication efficiency, the energy implications of federated learning (FL) within Artificial Intelligence of Things (AIoT) scenarios are often overlooked in the existing literature. This study examines the energy consumed during the FL process, focusing on three main energy-intensive processes: 1) preprocessing; 2) communication; and 3) local learning, all contributing to the overall energy footprint. We rely on the observation that device/client selection is crucial for speeding up the convergence of model training in a distributed AIoT setting and propose two clustering-informed methods. These clustering solutions are designed to group AIoT devices with similar label distributions, resulting in clusters composed of nearly heterogeneous devices. Hence, our methods alleviate the heterogeneity often encountered in real-world distributed learning applications. Throughout extensive numerical experimentation, we demonstrate that our clustering strategies typically achieve high convergence rates while maintaining low energy consumption when compared to other recent approaches available in the literature.Institute of Electrical and Electronics Engineers Inc.2025info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttps://cttc.fundanetsuite.com/Publicaciones/ProdCientif/PublicacionFrw.aspx?id=8699IEEE Internet of Things JournalISSN: 23274662reponame:r-CTTC. Repositorio Institucional Producción Científica del Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)instname:Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)Inglésinfo:eu-repo/semantics/openAccessoai:cttc.fundanetsuite.com:p86992026-06-17T11:44:47Z
dc.title.none.fl_str_mv Energy-Efficient Federated Learning for AIoT Using Clustering Methods
title Energy-Efficient Federated Learning for AIoT Using Clustering Methods
spellingShingle Energy-Efficient Federated Learning for AIoT Using Clustering Methods
Pereira R.
Training
Internet of Things
Artificial intelligence
Convergence
Data models
Servers
Computational modeling
Costs
Distributed databases
Energy consumption
Artificial Intelligence of Things (AIoT) device selection
clustering
energy efficiency
federated learning (FL)
title_short Energy-Efficient Federated Learning for AIoT Using Clustering Methods
title_full Energy-Efficient Federated Learning for AIoT Using Clustering Methods
title_fullStr Energy-Efficient Federated Learning for AIoT Using Clustering Methods
title_full_unstemmed Energy-Efficient Federated Learning for AIoT Using Clustering Methods
title_sort Energy-Efficient Federated Learning for AIoT Using Clustering Methods
dc.creator.none.fl_str_mv Pereira R.
Fama F.
Kalalas C.
Dini P.
author Pereira R.
author_facet Pereira R.
Fama F.
Kalalas C.
Dini P.
author_role author
author2 Fama F.
Kalalas C.
Dini P.
author2_role author
author
author
dc.subject.none.fl_str_mv Training
Internet of Things
Artificial intelligence
Convergence
Data models
Servers
Computational modeling
Costs
Distributed databases
Energy consumption
Artificial Intelligence of Things (AIoT) device selection
clustering
energy efficiency
federated learning (FL)
topic Training
Internet of Things
Artificial intelligence
Convergence
Data models
Servers
Computational modeling
Costs
Distributed databases
Energy consumption
Artificial Intelligence of Things (AIoT) device selection
clustering
energy efficiency
federated learning (FL)
description While substantial research has been devoted to optimizing model performance, convergence rates, and communication efficiency, the energy implications of federated learning (FL) within Artificial Intelligence of Things (AIoT) scenarios are often overlooked in the existing literature. This study examines the energy consumed during the FL process, focusing on three main energy-intensive processes: 1) preprocessing; 2) communication; and 3) local learning, all contributing to the overall energy footprint. We rely on the observation that device/client selection is crucial for speeding up the convergence of model training in a distributed AIoT setting and propose two clustering-informed methods. These clustering solutions are designed to group AIoT devices with similar label distributions, resulting in clusters composed of nearly heterogeneous devices. Hence, our methods alleviate the heterogeneity often encountered in real-world distributed learning applications. Throughout extensive numerical experimentation, we demonstrate that our clustering strategies typically achieve high convergence rates while maintaining low energy consumption when compared to other recent approaches available in the literature.
publishDate 2025
dc.date.none.fl_str_mv 2025
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://cttc.fundanetsuite.com/Publicaciones/ProdCientif/PublicacionFrw.aspx?id=8699
url https://cttc.fundanetsuite.com/Publicaciones/ProdCientif/PublicacionFrw.aspx?id=8699
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers Inc.
publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers Inc.
dc.source.none.fl_str_mv IEEE Internet of Things Journal
ISSN: 23274662
reponame:r-CTTC. Repositorio Institucional Producción Científica del Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
instname:Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
instname_str Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
reponame_str r-CTTC. Repositorio Institucional Producción Científica del Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
collection r-CTTC. Repositorio Institucional Producción Científica del Centre Tecnològic de Telecomunicacions de Catalunya (CTTC)
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869416306037489664
score 15,811543

Similares en LA Referencia