Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT
<p dir="ltr">Aiming at protecting device data privacy, Federated Learning (FL) is a framework of distributed machine learning in which devices’ local model parameters are exchanged with a centralized server without revealing the actual data. Hierarchical Federated Learning (HFL) fram...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , |
| منشور في: |
2023
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1864513535652921344 |
|---|---|
| author | Noora Mohammed Al-Maslamani (17541798) |
| author2 | Mohamed Abdallah (3073191) Bekir Sait Ciftler (17541801) |
| author2_role | author author |
| author_facet | Noora Mohammed Al-Maslamani (17541798) Mohamed Abdallah (3073191) Bekir Sait Ciftler (17541801) |
| author_role | author |
| dc.creator.none.fl_str_mv | Noora Mohammed Al-Maslamani (17541798) Mohamed Abdallah (3073191) Bekir Sait Ciftler (17541801) |
| dc.date.none.fl_str_mv | 2023-06-01T03:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1109/ojcoms.2023.3280359 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Reputation-Aware_Multi-Agent_DRL_for_Secure_Hierarchical_Federated_Learning_in_IoT/24717390 |
| 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 Servers Image edge detection Training Optimization Internet of Things Data models Reinforcement learning |
| dc.title.none.fl_str_mv | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Aiming at protecting device data privacy, Federated Learning (FL) is a framework of distributed machine learning in which devices’ local model parameters are exchanged with a centralized server without revealing the actual data. Hierarchical Federated Learning (HFL) framework was introduced to improve FL communication efficiency where devices are clustered and seek model consensus with the support of edge servers (e.g., base stations). Devices in a cluster submit their local model updates to their assigned local edge server for aggregation at each iteration. The edge servers transmit the aggregated models to a centralized server and establish a global consensus. However, similar to FL, adversaries may threaten the security and privacy of HFL. The client devices within a cluster may deliberately provide unreliable local model updates through poisoning attacks or poor-quality model updates due to inconsistent communication channels, increased device mobility, or inadequate device resources. To address the above challenges, this paper investigates the client selection problem in the HFL framework to eliminate the impact of unreliable clients while maximizing the global model accuracy of HFL. Each FL edge server is equipped with a Deep Reinforcement Learning (DRL)-based reputation model to optimally measure the reliability and trustworthiness of FL workers within its cluster. A Multi-Agent Deep Deterministic Policy Gradient (MADDPG) is utilized to enhance the accuracy and stability of the HFL global model, given the workers’ dynamic behaviors in the HFL environment. The experimental results indicate that our proposed MADDPG improves the accuracy and stability of HFL compared with the conventional reputation model and single-agent DDPG-based reputation model.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Communications Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcoms.2023.3280359" target="_blank">https://dx.doi.org/10.1109/ojcoms.2023.3280359</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_006465732bbd28066c5e129f9052f62f |
| identifier_str_mv | 10.1109/ojcoms.2023.3280359 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/24717390 |
| publishDate | 2023 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoTNoora Mohammed Al-Maslamani (17541798)Mohamed Abdallah (3073191)Bekir Sait Ciftler (17541801)Information and computing sciencesDistributed computing and systems softwareMachine learningServersImage edge detectionTrainingOptimizationInternet of ThingsData modelsReinforcement learning<p dir="ltr">Aiming at protecting device data privacy, Federated Learning (FL) is a framework of distributed machine learning in which devices’ local model parameters are exchanged with a centralized server without revealing the actual data. Hierarchical Federated Learning (HFL) framework was introduced to improve FL communication efficiency where devices are clustered and seek model consensus with the support of edge servers (e.g., base stations). Devices in a cluster submit their local model updates to their assigned local edge server for aggregation at each iteration. The edge servers transmit the aggregated models to a centralized server and establish a global consensus. However, similar to FL, adversaries may threaten the security and privacy of HFL. The client devices within a cluster may deliberately provide unreliable local model updates through poisoning attacks or poor-quality model updates due to inconsistent communication channels, increased device mobility, or inadequate device resources. To address the above challenges, this paper investigates the client selection problem in the HFL framework to eliminate the impact of unreliable clients while maximizing the global model accuracy of HFL. Each FL edge server is equipped with a Deep Reinforcement Learning (DRL)-based reputation model to optimally measure the reliability and trustworthiness of FL workers within its cluster. A Multi-Agent Deep Deterministic Policy Gradient (MADDPG) is utilized to enhance the accuracy and stability of the HFL global model, given the workers’ dynamic behaviors in the HFL environment. The experimental results indicate that our proposed MADDPG improves the accuracy and stability of HFL compared with the conventional reputation model and single-agent DDPG-based reputation model.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Communications Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcoms.2023.3280359" target="_blank">https://dx.doi.org/10.1109/ojcoms.2023.3280359</a></p>2023-06-01T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/ojcoms.2023.3280359https://figshare.com/articles/journal_contribution/Reputation-Aware_Multi-Agent_DRL_for_Secure_Hierarchical_Federated_Learning_in_IoT/24717390CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/247173902023-06-01T03:00:00Z |
| spellingShingle | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT Noora Mohammed Al-Maslamani (17541798) Information and computing sciences Distributed computing and systems software Machine learning Servers Image edge detection Training Optimization Internet of Things Data models Reinforcement learning |
| status_str | publishedVersion |
| title | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| title_full | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| title_fullStr | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| title_full_unstemmed | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| title_short | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| title_sort | Reputation-Aware Multi-Agent DRL for Secure Hierarchical Federated Learning in IoT |
| topic | Information and computing sciences Distributed computing and systems software Machine learning Servers Image edge detection Training Optimization Internet of Things Data models Reinforcement learning |