Intelligent Bilateral Client Selection in Federated Learning Using Game Theory

Federated Learning (FL) is a novel distributed privacy-preserving learning paradigm, which enables the collaboration among several participants (e.g., Internet of Things devices) for the training of machine learning models. However, selecting the participants that would contribute to this collaborat...

Full description

Saved in:
Bibliographic Details
Main Author: Wehbi, Osama (author)
Format: masterThesis
Published: 2022
Subjects:
Online Access:http://hdl.handle.net/10725/14123
https://doi.org/10.26756/th.2022.451
http://libraries.lau.edu.lb/research/laur/terms-of-use/thesis.php
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513468818784256
author Wehbi, Osama
author_facet Wehbi, Osama
author_role author
dc.creator.none.fl_str_mv Wehbi, Osama
dc.date.none.fl_str_mv 2022-10-25T07:02:52Z
2022-10-25T07:02:52Z
2022
2022-08-18
dc.identifier.none.fl_str_mv http://hdl.handle.net/10725/14123
https://doi.org/10.26756/th.2022.451
http://libraries.lau.edu.lb/research/laur/terms-of-use/thesis.php
dc.language.none.fl_str_mv en
dc.publisher.none.fl_str_mv Lebanese American University
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Machine learning -- Case studies
Game theory
Internet of things
Computational intelligence
Data privacy
Lebanese American University -- Dissertations
Dissertations, Academic
dc.title.none.fl_str_mv Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
dc.type.none.fl_str_mv Thesis
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/masterThesis
description Federated Learning (FL) is a novel distributed privacy-preserving learning paradigm, which enables the collaboration among several participants (e.g., Internet of Things devices) for the training of machine learning models. However, selecting the participants that would contribute to this collaborative training is highly challenging. Adopting a random selection strategy would entail substantial problems due to the heterogeneity in terms of data quality, and computational and communication resources across the participants. To overcome this problem, we present in this paper FedMint, an intelligent client selection approach for federated learning on IoT devices using game theory and bootstrapping mechanism. Our solution involves designing (1) preference functions for the client IoT devices and federated servers to allow them to rank each other according to several factors such as accuracy and price, (2) intelligent matching algorithms that take into account the preferences of both parties in their design, and (3) bootstrapping technique that capitalizes on the collaboration of multiple federated servers in order to assign initial accuracy value for the new connected IoT devices. Based on our simulation findings, our strategy surpasses the VanillaF selection approach in terms of maximizing both the revenues of the client devices and accuracy of the global federated learning model.
eu_rights_str_mv openAccess
format masterThesis
id LAURepo_c4ab11c9a9dbfcc96effc60e5d76973a
language_invalid_str_mv en
network_acronym_str LAURepo
network_name_str Lebanese American University repository
oai_identifier_str oai:laur.lau.edu.lb:10725/14123
publishDate 2022
publisher.none.fl_str_mv Lebanese American University
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Intelligent Bilateral Client Selection in Federated Learning Using Game TheoryWehbi, OsamaMachine learning -- Case studiesGame theoryInternet of thingsComputational intelligenceData privacyLebanese American University -- DissertationsDissertations, AcademicFederated Learning (FL) is a novel distributed privacy-preserving learning paradigm, which enables the collaboration among several participants (e.g., Internet of Things devices) for the training of machine learning models. However, selecting the participants that would contribute to this collaborative training is highly challenging. Adopting a random selection strategy would entail substantial problems due to the heterogeneity in terms of data quality, and computational and communication resources across the participants. To overcome this problem, we present in this paper FedMint, an intelligent client selection approach for federated learning on IoT devices using game theory and bootstrapping mechanism. Our solution involves designing (1) preference functions for the client IoT devices and federated servers to allow them to rank each other according to several factors such as accuracy and price, (2) intelligent matching algorithms that take into account the preferences of both parties in their design, and (3) bootstrapping technique that capitalizes on the collaboration of multiple federated servers in order to assign initial accuracy value for the new connected IoT devices. Based on our simulation findings, our strategy surpasses the VanillaF selection approach in terms of maximizing both the revenues of the client devices and accuracy of the global federated learning model.1 online resource (x, 45 leaves): col. ill.Bibliography: leaves 42-45.Lebanese American University2022-10-25T07:02:52Z2022-10-25T07:02:52Z20222022-08-18Thesisinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesishttp://hdl.handle.net/10725/14123https://doi.org/10.26756/th.2022.451http://libraries.lau.edu.lb/research/laur/terms-of-use/thesis.phpeninfo:eu-repo/semantics/openAccessoai:laur.lau.edu.lb:10725/141232022-10-25T07:03:59Z
spellingShingle Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
Wehbi, Osama
Machine learning -- Case studies
Game theory
Internet of things
Computational intelligence
Data privacy
Lebanese American University -- Dissertations
Dissertations, Academic
status_str publishedVersion
title Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
title_full Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
title_fullStr Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
title_full_unstemmed Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
title_short Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
title_sort Intelligent Bilateral Client Selection in Federated Learning Using Game Theory
topic Machine learning -- Case studies
Game theory
Internet of things
Computational intelligence
Data privacy
Lebanese American University -- Dissertations
Dissertations, Academic
url http://hdl.handle.net/10725/14123
https://doi.org/10.26756/th.2022.451
http://libraries.lau.edu.lb/research/laur/terms-of-use/thesis.php