Knowledge- and Model-Driven Deep Reinforcement Learning for Efficient Federated Edge Learning: Single- and Multi-Agent Frameworks

In this paper, we investigate federated learning (FL) efficiency improvement in practical edge computing systems, where edge workers have non-independent and identically distributed (non-IID) local data, as well as dynamic and heterogeneous computing and communication capabilities. We consider a gen...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on machine learning in communications and networking Jg. 3; S. 332 - 352
Hauptverfasser: Li, Yangchen, Zhao, Lingzhi, Wang, Tianle, Ding, Lianghui, Yang, Feng
Format: Journal Article
Sprache:Englisch
Veröffentlicht: IEEE 2025
Schlagworte:
ISSN:2831-316X, 2831-316X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract In this paper, we investigate federated learning (FL) efficiency improvement in practical edge computing systems, where edge workers have non-independent and identically distributed (non-IID) local data, as well as dynamic and heterogeneous computing and communication capabilities. We consider a general FL algorithm with configurable parameters, including the number of local iterations, mini-batch sizes, step sizes, aggregation weights, and quantization parameters, and provide a rigorous convergence analysis. We formulate a joint optimization problem for FL worker selection and algorithm parameter configuration to minimize the final test loss subject to time and energy constraints. The resulting problem is a complicated stochastic sequential decision-making problem with an implicit objective function and unknown transition probabilities. To address these challenges, we propose knowledge/model-driven single-agent and multi-agent deep reinforcement learning (DRL) frameworks. We transform the primal problem into a Markov decision process (MDP) for the single-agent DRL framework and a decentralized partially-observable Markov decision process (Dec-POMDP) for the multi-agent DRL framework. We develop efficient single-agent and multi-agent asynchronous advantage actor-critic (A3C) approaches to solve the MDP and Dec-POMDP, respectively. In both frameworks, we design a knowledge-based reward to facilitate effective DRL and propose a model-based stochastic policy to tackle the mixed discrete-continuous actions and large action spaces. To reduce the computational complexities of policy learning and execution, we introduce a segmented actor-critic architecture for the single-agent DRL and a distributed actor-critic architecture for the multi-agent DRL. Numerical results demonstrate the effectiveness and advantages of the proposed frameworks in enhancing FL efficiency.
AbstractList In this paper, we investigate federated learning (FL) efficiency improvement in practical edge computing systems, where edge workers have non-independent and identically distributed (non-IID) local data, as well as dynamic and heterogeneous computing and communication capabilities. We consider a general FL algorithm with configurable parameters, including the number of local iterations, mini-batch sizes, step sizes, aggregation weights, and quantization parameters, and provide a rigorous convergence analysis. We formulate a joint optimization problem for FL worker selection and algorithm parameter configuration to minimize the final test loss subject to time and energy constraints. The resulting problem is a complicated stochastic sequential decision-making problem with an implicit objective function and unknown transition probabilities. To address these challenges, we propose knowledge/model-driven single-agent and multi-agent deep reinforcement learning (DRL) frameworks. We transform the primal problem into a Markov decision process (MDP) for the single-agent DRL framework and a decentralized partially-observable Markov decision process (Dec-POMDP) for the multi-agent DRL framework. We develop efficient single-agent and multi-agent asynchronous advantage actor-critic (A3C) approaches to solve the MDP and Dec-POMDP, respectively. In both frameworks, we design a knowledge-based reward to facilitate effective DRL and propose a model-based stochastic policy to tackle the mixed discrete-continuous actions and large action spaces. To reduce the computational complexities of policy learning and execution, we introduce a segmented actor-critic architecture for the single-agent DRL and a distributed actor-critic architecture for the multi-agent DRL. Numerical results demonstrate the effectiveness and advantages of the proposed frameworks in enhancing FL efficiency.
Author Ding, Lianghui
Li, Yangchen
Zhao, Lingzhi
Yang, Feng
Wang, Tianle
Author_xml – sequence: 1
  givenname: Yangchen
  orcidid: 0000-0001-5885-5587
  surname: Li
  fullname: Li, Yangchen
  organization: Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China
– sequence: 2
  givenname: Lingzhi
  orcidid: 0000-0002-8104-1124
  surname: Zhao
  fullname: Zhao, Lingzhi
  organization: Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA
– sequence: 3
  givenname: Tianle
  surname: Wang
  fullname: Wang, Tianle
  organization: Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China
– sequence: 4
  givenname: Lianghui
  orcidid: 0000-0002-3231-3613
  surname: Ding
  fullname: Ding, Lianghui
  organization: Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China
– sequence: 5
  givenname: Feng
  orcidid: 0000-0002-6350-5765
  surname: Yang
  fullname: Yang, Feng
  email: yangfeng@sjtu.edu.cn
  organization: Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China
BookMark eNp9kc1OGzEURq2KSqWBF6i68AtM6t_xuDsUQosIVAIqdWfdse9EphMbeYaiLvvmTBKoUBfd2NZnn09XPu_JQcoJCfnA2ZxzZj_dXq4WV3PBhJ5LLZXR6g05FI3kleT1j4NX53fkeBjuGGNSMmNqcUj-XKT82GNYY0UhBXqZA_bVaYm_MNFTxHt6jTF1uXjcYBrpCqGkmNZ0iuiy66KP2_gMAxYYMdDlVPX31Wd6M639S_dDP8bqZL0DCmzwMZefwxF520E_4PHzPiPfz5a3i6_V6tuX88XJqvKSNWNlhO5a7VlbdyJw0J03nCndmq4xtlUIjWfA2HSBlhthTfASrIGaeyu9lHJGzve9IcOduy9xA-W3yxDdLshl7aCM0ffoZF2D0FoZFbzS0luuGtaitdzXAaY_npFm3-VLHoaCnfNxhDHmNBaIvePMbc24nRm3NeOezUyo-Ad9GeW_0Mc9FBHxFdBopSeZTwgVnF0
CODEN ITMLBB
CitedBy_id crossref_primary_10_1109_TCC_2025_3574823
Cites_doi 10.1109/TII.2022.3210008
10.1109/5.726791
10.1109/JIOT.2022.3172113
10.1145/3545008.3545085
10.1109/JSTSP.2021.3126174
10.1109/TMC.2022.3186936
10.1109/TII.2021.3064351
10.1109/MWC.003.2100028
10.1109/JSAC.2019.2904348
10.1109/TWC.2020.3037554
10.1109/IWQoS49365.2020.9212881
10.1109/TWC.2020.3002988
10.1609/aaai.v36i8.20894
10.1109/SEC50012.2020.00060
10.1109/TNSE.2022.3225444
10.1109/GCWkshps50303.2020.9367421
10.1109/TWC.2016.2633522
10.1109/TPDS.2021.3134647
10.1561/2200000083
10.1109/TWC.2023.3335302
10.1109/JSAC.2021.3118352
10.1109/LCOMM.2021.3122129
10.1109/JIOT.2020.3028742
10.1109/COMST.2020.2986024
10.1016/j.future.2023.03.009
10.1109/FMEC59375.2023.10306017
10.1109/TCOMM.2021.3083316
10.1109/INFOCOM41043.2020.9155494
10.1109/IJCNN.2017.7966217
10.1109/INFOCOM.2019.8737464
10.1109/IWQOS52092.2021.9521301
10.1109/TII.2022.3222369
10.1109/ICDE53745.2022.00077
10.1109/INFOCOM42981.2021.9488679
10.1109/GLOBECOM48099.2022.10001466
10.1109/TII.2022.3183465
10.1609/aaai.v33i01.33015693
10.23919/EUSIPCO54536.2021.9616052
10.1007/s11280-021-00989-x
10.1109/CVPR52688.2022.00986
10.1109/TC.2021.3099723
10.1109/ICC.2019.8761315
10.1109/IJCNN55064.2022.9892211
10.1007/978-3-319-71682-4_5
10.1109/ACCESS.2021.3056919
ContentType Journal Article
DBID 97E
ESBDL
RIA
RIE
AAYXX
CITATION
DOA
DOI 10.1109/TMLCN.2025.3534754
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE Xplore Open Access Journals
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Explore
CrossRef
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
DatabaseTitleList

Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: RIE
  name: IEEE Explore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 2831-316X
EndPage 352
ExternalDocumentID oai_doaj_org_article_366a255474dc453c91480be991c6da35
10_1109_TMLCN_2025_3534754
10854500
Genre orig-research
GrantInformation_xml – fundername: Shanghai Key Laboratory Funding
  grantid: (STCSM 22DZ2229005)
– fundername: Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning
  funderid: 10.13039/501100013285
GroupedDBID 0R~
97E
ABVLG
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
EBS
ESBDL
GROUPED_DOAJ
JAVBF
M~E
OCL
RIA
RIE
AAYXX
CITATION
ID FETCH-LOGICAL-c308t-725fb5c0b6f2d1a5fc71045b7f879b4ea8c0a00a5fe917297dc3a97a61c93c333
IEDL.DBID RIE
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001487810100003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2831-316X
IngestDate Fri Oct 03 12:45:03 EDT 2025
Sat Nov 29 08:18:42 EST 2025
Tue Nov 18 22:17:19 EST 2025
Wed Aug 27 01:49:24 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Language English
License https://creativecommons.org/licenses/by/4.0/legalcode
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c308t-725fb5c0b6f2d1a5fc71045b7f879b4ea8c0a00a5fe917297dc3a97a61c93c333
ORCID 0000-0001-5885-5587
0000-0002-3231-3613
0000-0002-6350-5765
0000-0002-8104-1124
OpenAccessLink https://ieeexplore.ieee.org/document/10854500
PageCount 21
ParticipantIDs ieee_primary_10854500
doaj_primary_oai_doaj_org_article_366a255474dc453c91480be991c6da35
crossref_citationtrail_10_1109_TMLCN_2025_3534754
crossref_primary_10_1109_TMLCN_2025_3534754
PublicationCentury 2000
PublicationDate 20250000
2025-00-00
2025-01-01
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – year: 2025
  text: 20250000
PublicationDecade 2020
PublicationTitle IEEE transactions on machine learning in communications and networking
PublicationTitleAbbrev TMLCN
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
References ref13
ref57
ref12
ref15
ref14
ref53
ref11
Li (ref6) 2020; 2
ref54
ref17
Li (ref1)
ref16
ref19
ref18
Mnih (ref46)
ref45
ref48
Schulman (ref56) 2017
ref47
ref42
ref41
ref44
ref43
ref49
ref8
ref9
ref3
ref5
Kingma (ref52) 2014
ref40
ref35
ref34
ref37
ref36
ref31
ref30
ref33
McMahan (ref4); 54
ref32
ref2
ref39
ref38
Krizhevsky (ref55) 2009
ref24
Bertsekas (ref51) 2019; 1
ref26
ref25
ref20
ref22
ref21
ref28
Jee Cho (ref23) 2020
ref27
Hsu (ref7) 2019
ref29
Wang (ref10); 1
Zhao (ref50) 2018
References_xml – ident: ref37
  doi: 10.1109/TII.2022.3210008
– start-page: 1928
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref46
  article-title: Asynchronous methods for deep reinforcement learning
– ident: ref45
  doi: 10.1109/5.726791
– ident: ref19
  doi: 10.1109/JIOT.2022.3172113
– ident: ref32
  doi: 10.1145/3545008.3545085
– ident: ref26
  doi: 10.1109/JSTSP.2021.3126174
– ident: ref9
  doi: 10.1109/TMC.2022.3186936
– ident: ref39
  doi: 10.1109/TII.2021.3064351
– ident: ref5
  doi: 10.1109/MWC.003.2100028
– ident: ref12
  doi: 10.1109/JSAC.2019.2904348
– year: 2017
  ident: ref56
  article-title: Proximal policy optimization algorithms
  publication-title: arXiv:1707.06347
– ident: ref15
  doi: 10.1109/TWC.2020.3037554
– ident: ref29
  doi: 10.1109/IWQoS49365.2020.9212881
– year: 2019
  ident: ref7
  article-title: Measuring the effects of non-identical data distribution for federated visual classification
  publication-title: arXiv:1909.06335
– ident: ref13
  doi: 10.1109/TWC.2020.3002988
– ident: ref35
  doi: 10.1609/aaai.v36i8.20894
– start-page: 1
  volume-title: Proc. IEEE Global Commun. Conf. (Globecom)
  ident: ref1
  article-title: Knowledge and model-driven deep reinforcement learning for federated edge learning
– ident: ref17
  doi: 10.1109/SEC50012.2020.00060
– ident: ref42
  doi: 10.1109/TNSE.2022.3225444
– ident: ref28
  doi: 10.1109/GCWkshps50303.2020.9367421
– ident: ref48
  doi: 10.1109/TWC.2016.2633522
– volume: 1
  start-page: 212
  volume-title: Proc. SysML
  ident: ref10
  article-title: Adaptive communication strategies to achieve the best error-runtime trade-off in local-update SGD
– ident: ref43
  doi: 10.1109/TPDS.2021.3134647
– ident: ref49
  doi: 10.1561/2200000083
– volume: 1
  year: 2019
  ident: ref51
  publication-title: Reinforcement Learning and Optimal Control
– ident: ref11
  doi: 10.1109/TWC.2023.3335302
– ident: ref40
  doi: 10.1109/JSAC.2021.3118352
– ident: ref44
  doi: 10.1109/LCOMM.2021.3122129
– ident: ref27
  doi: 10.1109/JIOT.2020.3028742
– ident: ref2
  doi: 10.1109/COMST.2020.2986024
– ident: ref41
  doi: 10.1016/j.future.2023.03.009
– ident: ref3
  doi: 10.1109/FMEC59375.2023.10306017
– ident: ref8
  doi: 10.1109/TCOMM.2021.3083316
– ident: ref34
  doi: 10.1109/INFOCOM41043.2020.9155494
– ident: ref54
  doi: 10.1109/IJCNN.2017.7966217
– ident: ref14
  doi: 10.1109/INFOCOM.2019.8737464
– ident: ref21
  doi: 10.1109/IWQOS52092.2021.9521301
– year: 2014
  ident: ref52
  article-title: Adam: A method for stochastic optimization
  publication-title: arXiv:1412.6980
– volume-title: Learning multiple layers of features from tiny images
  year: 2009
  ident: ref55
– ident: ref33
  doi: 10.1109/TII.2022.3222369
– volume: 54
  start-page: 1273
  volume-title: Proc. 20th Int. Conf. Artif. Intell. Statist.
  ident: ref4
  article-title: Communication-efficient learning of deep networks from decentralized data
– ident: ref57
  doi: 10.1109/ICDE53745.2022.00077
– ident: ref25
  doi: 10.1109/INFOCOM42981.2021.9488679
– ident: ref38
  doi: 10.1109/GLOBECOM48099.2022.10001466
– ident: ref36
  doi: 10.1109/TII.2022.3183465
– ident: ref47
  doi: 10.1609/aaai.v33i01.33015693
– ident: ref22
  doi: 10.23919/EUSIPCO54536.2021.9616052
– ident: ref16
  doi: 10.1007/s11280-021-00989-x
– ident: ref18
  doi: 10.1109/CVPR52688.2022.00986
– volume: 2
  start-page: 429
  year: 2020
  ident: ref6
  article-title: Federated optimization in heterogeneous networks
  publication-title: Proc. Mach. Learn. Syst.
– ident: ref31
  doi: 10.1109/TC.2021.3099723
– ident: ref24
  doi: 10.1109/ICC.2019.8761315
– ident: ref30
  doi: 10.1109/IJCNN55064.2022.9892211
– ident: ref53
  doi: 10.1007/978-3-319-71682-4_5
– ident: ref20
  doi: 10.1109/ACCESS.2021.3056919
– year: 2018
  ident: ref50
  article-title: Federated learning with non-IID data
  publication-title: arXiv:1806.00582
– year: 2020
  ident: ref23
  article-title: Client selection in federated learning: Convergence analysis and power-of-choice selection strategies
  publication-title: arXiv:2010.01243
SSID ssj0003307762
Score 2.2853618
Snippet In this paper, we investigate federated learning (FL) efficiency improvement in practical edge computing systems, where edge workers have non-independent and...
SourceID doaj
crossref
ieee
SourceType Open Website
Enrichment Source
Index Database
Publisher
StartPage 332
SubjectTerms Accuracy
algorithm parameter configuration
Computational modeling
Convergence
deep reinforcement learning
Delay effects
Edge computing
Energy consumption
Federated learning
Markov decision processes
multi-agent reinforcement learning
Quantization (signal)
Servers
Stochastic processes
worker selection
SummonAdditionalLinks – databaseName: DOAJ Directory of Open Access Journals
  dbid: DOA
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LS8QwEA4iHvQgPnF9kYM3qWabtGm8-dhFUBfxAd5KMklFWFZZV-_-c2fSdlkvevHSQ0jSkJkk34TJ9zF2QJjUk3x7ATkFKBISVzniIgzgQiBCb4hiE3owKJ6ezO2M1BflhNX0wPXEHcs8twh7lVYeVCbBIH4XLiCsgdxbGdlLEfXMBFO0B2OUrnGZt69khDl-uLk-H2A8mGZHMpNKZ-rHSRQJ-38orMQDpr_ClhtkyE_rEa2yuTBaY0szfIHr7OuqvQFLuB15TkJmw-RiTDsWvwjhjd-FyIQK8dKPN-SpzxyLeC-SRVBxnwgkEGN63sOuprVO-D1-h23flGmYnD7HBm0G1_sGe-z3Hs4vk0ZDIQEpikmi06xyGQiXV6nv2qwChBQqc7oqtHEq2AKEFcJSzhliGaM9SGu0zbtgJEgpN9n86HUUthgnsjofRI5BEKg04NJNU6u8U65IgwPTYd12PktoCMZJ52JYxkBDmDLaoCQblI0NOuxw2uatptf4tfYZmWlak6ixYwE6TNk4TPmXw3TYBhl55ncFokghtv-j8x22SAOu72h22fxk_BH22AJ8Tl7ex_vRR78BSlvm3g
  priority: 102
  providerName: Directory of Open Access Journals
Title Knowledge- and Model-Driven Deep Reinforcement Learning for Efficient Federated Edge Learning: Single- and Multi-Agent Frameworks
URI https://ieeexplore.ieee.org/document/10854500
https://doaj.org/article/366a255474dc453c91480be991c6da35
Volume 3
WOSCitedRecordID wos001487810100003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2831-316X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0003307762
  issn: 2831-316X
  databaseCode: DOA
  dateStart: 20230101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2831-316X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0003307762
  issn: 2831-316X
  databaseCode: M~E
  dateStart: 20230101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Na9wwEBVp6KE9pF8p3TYNOvRWnGgtybJ6S5NdCk2WkqSQm5FG41BYNmGzybHQf94Z2V6SQwq9CDNIsvBY1rzxzBshPrFNmrh8ew0VAxQNRWwjcxEiREQm9IZcbMLNZvXFhf_RJ6vnXBhEzMFnuMeX-V9-uoJbdpXtc6S8sYoQ-hPnXJestXaoEDB3tLOHxBjl989Pjg9nBAFLu6etNs6aB4dP5uh_UFQlnynTF_-5mpdiqzce5UGn7VdiAxevxfN7lIJvxJ_vg5OskGGRJNc6mxdHS_6oySPEa3mKmSwVsl9Q9vyql5JEcpL5JFg8ZY4JMkOTnNBU615f5Bm182FuDkYsDi7zgCHI62Zb_JxOzg-_FX2ZhQK0qleFK20bLahYtWUaB9sCWR3GRtfWzkeDoQYVlAoclkbmjncJdPAuVGPwGrTWb8Xm4mqB74RkPruEqiKcBKZE2t1lGUyKJtYlRvAjMR6efwM9BzmXwpg3GYso32SdNayzptfZSHxej7nuGDj-2fsrq3Xdk9mzs4BU1_SbsdFVFQhKGWcSGKvBEyZUEclUhioFbUdim9V973adpt8_Iv8gnvEaOs_MjthcLW_xo3gKd6tfN8vdDPOpPfk92c2v7F-wr-eX
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Na9wwEBUlLbQ9pF8J3fRLh96KE60lWVZvabJLStKltCnkZqTROBSWTdhseu8_74xsL8khhVyMEZIs_Cx73njmjRAf2SZNXL69hooJioYitpG1CBEiIgt6Qy424Waz-uzMf--T1XMuDCLm4DPc5dP8Lz9dwDW7yvY4Ut5YRQz9oTWmHHfpWmuXClFzR3t7SI1Rfu_028nBjEhgaXe11cZZc-vzk1X6b5VVyV-V6bN7rue52OzNR7nf4f1CPMDFS_H0hqjgK_H3eHCTFTIskuRqZ_PicMmvNXmIeCl_YJZLhewZlL3C6rmkJjnJihLcPGWVCTJEk5zQVOten-VPOs6HuTkcsdg_zwOGMK-rLfFrOjk9OCr6QgsFaFWvClfaNlpQsWrLNA62BbI7jI2urZ2PBkMNKigVODCNDB7vEujgXajG4DVorbfFxuJiga-FZEW7hKoipgSmRNrfZRlMiibWJUbwIzEe7n8DvQo5F8OYN5mNKN9kzBrGrOkxG4lP6zGXnQbHf3t_YVjXPVk_OzcQdE2_HRtdVYHIlHEmgbEaPLFCFZGMZahS0HYkthjuG5frkN65o_2DeHxES2lOvs6O34gnvJ7OT_NWbKyW1_hOPII_q99Xy_f5kf0H9izouA
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Knowledge-+and+Model-Driven+Deep+Reinforcement+Learning+for+Efficient+Federated+Edge+Learning%3A+Single-+and+Multi-Agent+Frameworks&rft.jtitle=IEEE+transactions+on+machine+learning+in+communications+and+networking&rft.au=Li%2C+Yangchen&rft.au=Zhao%2C+Lingzhi&rft.au=Wang%2C+Tianle&rft.au=Ding%2C+Lianghui&rft.date=2025&rft.pub=IEEE&rft.eissn=2831-316X&rft.volume=3&rft.spage=332&rft.epage=352&rft_id=info:doi/10.1109%2FTMLCN.2025.3534754&rft.externalDocID=10854500
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2831-316X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2831-316X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2831-316X&client=summon