Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data

<p dir="ltr">Federated Learning (FL) is a distributed learning methodology that allows multiple nodes to cooperatively train a deep learning model, without the need to share their local data. It is a promising solution for telemonitoring systems that demand intensive data collection,...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Alaa Awad Abdellatif (17151163) (author)
مؤلفون آخرون: Naram Mhaisen (16870071) (author), Amr Mohamed (3508121) (author), Aiman Erbad (14150589) (author), Mohsen Guizani (12580291) (author), Zaher Dawy (17151166) (author), Wassim Nasreddine (9149936) (author)
منشور في: 2022
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513552559112192
author Alaa Awad Abdellatif (17151163)
author2 Naram Mhaisen (16870071)
Amr Mohamed (3508121)
Aiman Erbad (14150589)
Mohsen Guizani (12580291)
Zaher Dawy (17151166)
Wassim Nasreddine (9149936)
author2_role author
author
author
author
author
author
author_facet Alaa Awad Abdellatif (17151163)
Naram Mhaisen (16870071)
Amr Mohamed (3508121)
Aiman Erbad (14150589)
Mohsen Guizani (12580291)
Zaher Dawy (17151166)
Wassim Nasreddine (9149936)
author_role author
dc.creator.none.fl_str_mv Alaa Awad Abdellatif (17151163)
Naram Mhaisen (16870071)
Amr Mohamed (3508121)
Aiman Erbad (14150589)
Mohsen Guizani (12580291)
Zaher Dawy (17151166)
Wassim Nasreddine (9149936)
dc.date.none.fl_str_mv 2022-03-01T00:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.future.2021.10.016
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Communication-efficient_hierarchical_federated_learning_for_IoT_heterogeneous_systems_with_imbalanced_data/24314350
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Information and computing sciences
Distributed computing and systems software
Machine learning
Distributed deep learning
Edge computing
Non-IID data
Internet of Things (IoT)
Intelligent health systems
dc.title.none.fl_str_mv Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Federated Learning (FL) is a distributed learning methodology that allows multiple nodes to cooperatively train a deep learning model, without the need to share their local data. It is a promising solution for telemonitoring systems that demand intensive data collection, for detection, classification, and prediction of future events, from different locations while maintaining a strict privacy constraint. Due to privacy concerns and critical communication bottlenecks, it can become impractical to send the FL updated models to a centralized server. Thus, this paper studies the potential of hierarchical FL in Internet of Things (IoT) heterogeneous systems. In particular, we propose an optimized solution for user assignment and resource allocation over hierarchical FL architecture for IoT heterogeneous systems. This work focuses on a generic class of machine learning models that are trained using gradient-descent-based schemes while considering the practical constraints of non-uniformly distributed data across different users. We evaluate the proposed system using two real-world datasets, and we show that it outperforms state-of-the-art FL solutions. Specifically, our numerical results highlight the effectiveness of our approach and its ability to provide 4–6% increase in the classification accuracy, with respect to hierarchical FL schemes that consider distance-based user assignment. Furthermore, the proposed approach could significantly accelerate FL training and reduce communication overhead by providing 75–85% reduction in the communication rounds between edge nodes and the centralized server, for the same model accuracy.</p><h2>Other Information</h2><p dir="ltr">Published in: Future Generation Computer Systems<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.future.2021.10.016" target="_blank">https://dx.doi.org/10.1016/j.future.2021.10.016</a></p>
eu_rights_str_mv openAccess
id Manara2_3e21dd594cc803dbacbb33166eeef407
identifier_str_mv 10.1016/j.future.2021.10.016
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/24314350
publishDate 2022
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced dataAlaa Awad Abdellatif (17151163)Naram Mhaisen (16870071)Amr Mohamed (3508121)Aiman Erbad (14150589)Mohsen Guizani (12580291)Zaher Dawy (17151166)Wassim Nasreddine (9149936)Information and computing sciencesDistributed computing and systems softwareMachine learningDistributed deep learningEdge computingNon-IID dataInternet of Things (IoT)Intelligent health systems<p dir="ltr">Federated Learning (FL) is a distributed learning methodology that allows multiple nodes to cooperatively train a deep learning model, without the need to share their local data. It is a promising solution for telemonitoring systems that demand intensive data collection, for detection, classification, and prediction of future events, from different locations while maintaining a strict privacy constraint. Due to privacy concerns and critical communication bottlenecks, it can become impractical to send the FL updated models to a centralized server. Thus, this paper studies the potential of hierarchical FL in Internet of Things (IoT) heterogeneous systems. In particular, we propose an optimized solution for user assignment and resource allocation over hierarchical FL architecture for IoT heterogeneous systems. This work focuses on a generic class of machine learning models that are trained using gradient-descent-based schemes while considering the practical constraints of non-uniformly distributed data across different users. We evaluate the proposed system using two real-world datasets, and we show that it outperforms state-of-the-art FL solutions. Specifically, our numerical results highlight the effectiveness of our approach and its ability to provide 4–6% increase in the classification accuracy, with respect to hierarchical FL schemes that consider distance-based user assignment. Furthermore, the proposed approach could significantly accelerate FL training and reduce communication overhead by providing 75–85% reduction in the communication rounds between edge nodes and the centralized server, for the same model accuracy.</p><h2>Other Information</h2><p dir="ltr">Published in: Future Generation Computer Systems<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.future.2021.10.016" target="_blank">https://dx.doi.org/10.1016/j.future.2021.10.016</a></p>2022-03-01T00:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.future.2021.10.016https://figshare.com/articles/journal_contribution/Communication-efficient_hierarchical_federated_learning_for_IoT_heterogeneous_systems_with_imbalanced_data/24314350CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/243143502022-03-01T00:00:00Z
spellingShingle Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
Alaa Awad Abdellatif (17151163)
Information and computing sciences
Distributed computing and systems software
Machine learning
Distributed deep learning
Edge computing
Non-IID data
Internet of Things (IoT)
Intelligent health systems
status_str publishedVersion
title Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
title_full Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
title_fullStr Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
title_full_unstemmed Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
title_short Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
title_sort Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data
topic Information and computing sciences
Distributed computing and systems software
Machine learning
Distributed deep learning
Edge computing
Non-IID data
Internet of Things (IoT)
Intelligent health systems