FedPD: A Federated Learning Framework With Adaptivity to Non-IID Data
Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the cloud, algorithms such as the Federated Averaging (FedAvg) have adopted a computation then aggregation model, in which multiple local updates...
Saved in:
| Published in: | IEEE transactions on signal processing Vol. 69; pp. 6055 - 6070 |
|---|---|
| Main Authors: | , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
New York
IEEE
2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects: | |
| ISSN: | 1053-587X, 1941-0476 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the cloud, algorithms such as the Federated Averaging (FedAvg) have adopted a computation then aggregation model, in which multiple local updates are performed using local data before aggregation. These algorithms fail to work when faced with practical challenges, e.g., the local data being non-identically independently distributed. In this paper, we first characterize the behavior of the FedAvg algorithm, and show that without strong and unrealistic assumptions on the problem structure, it can behave erratically. Aiming at designing FL algorithms that are provably fast and require as few assumptions as possible, we propose a new algorithm design strategy from the primal-dual optimization perspective. Our strategy yields algorithms that can deal with non-convex objective functions, achieves the best possible optimization and communication complexity (in a well-defined sense), and accommodates full-batch and mini-batch local computation models. Importantly, the proposed algorithms are communication efficient , in that the communication effort can be reduced when the level of heterogeneity among the local data also reduces. In the extreme case where the local data becomes homogeneous, only <inline-formula><tex-math notation="LaTeX">\mathcal {O}(1)</tex-math></inline-formula> communication is required among the agents. To the best of our knowledge, this is the first algorithmic framework for FL that achieves all the above properties. |
|---|---|
| AbstractList | Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the cloud, algorithms such as the Federated Averaging (FedAvg) have adopted a computation then aggregation model, in which multiple local updates are performed using local data before aggregation. These algorithms fail to work when faced with practical challenges, e.g., the local data being non-identically independently distributed. In this paper, we first characterize the behavior of the FedAvg algorithm, and show that without strong and unrealistic assumptions on the problem structure, it can behave erratically. Aiming at designing FL algorithms that are provably fast and require as few assumptions as possible, we propose a new algorithm design strategy from the primal-dual optimization perspective. Our strategy yields algorithms that can deal with non-convex objective functions, achieves the best possible optimization and communication complexity (in a well-defined sense), and accommodates full-batch and mini-batch local computation models. Importantly, the proposed algorithms are communication efficient , in that the communication effort can be reduced when the level of heterogeneity among the local data also reduces. In the extreme case where the local data becomes homogeneous, only <inline-formula><tex-math notation="LaTeX">\mathcal {O}(1)</tex-math></inline-formula> communication is required among the agents. To the best of our knowledge, this is the first algorithmic framework for FL that achieves all the above properties. Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the cloud, algorithms such as the Federated Averaging (FedAvg) have adopted a computation then aggregation model, in which multiple local updates are performed using local data before aggregation. These algorithms fail to work when faced with practical challenges, e.g., the local data being non-identically independently distributed. In this paper, we first characterize the behavior of the FedAvg algorithm, and show that without strong and unrealistic assumptions on the problem structure, it can behave erratically. Aiming at designing FL algorithms that are provably fast and require as few assumptions as possible, we propose a new algorithm design strategy from the primal-dual optimization perspective. Our strategy yields algorithms that can deal with non-convex objective functions, achieves the best possible optimization and communication complexity (in a well-defined sense), and accommodates full-batch and mini-batch local computation models. Importantly, the proposed algorithms are communication efficient , in that the communication effort can be reduced when the level of heterogeneity among the local data also reduces. In the extreme case where the local data becomes homogeneous, only [Formula Omitted] communication is required among the agents. To the best of our knowledge, this is the first algorithmic framework for FL that achieves all the above properties. |
| Author | Dhople, Sairaj Liu, Yang Hong, Mingyi Zhang, Xinwei Yin, Wotao |
| Author_xml | – sequence: 1 givenname: Xinwei surname: Zhang fullname: Zhang, Xinwei email: zhan6234@umn.edu organization: University of Minnesota, Minneapolis, MN, USA – sequence: 2 givenname: Mingyi orcidid: 0000-0003-1263-9365 surname: Hong fullname: Hong, Mingyi email: mhong@umn.edu organization: University of Minnesota, Minneapolis, MN, USA – sequence: 3 givenname: Sairaj orcidid: 0000-0002-1180-1415 surname: Dhople fullname: Dhople, Sairaj email: sdhople@umn.edu organization: University of Minnesota, Minneapolis, MN, USA – sequence: 4 givenname: Wotao orcidid: 0000-0001-6697-9731 surname: Yin fullname: Yin, Wotao email: wotaoyin@math.ucla.edu organization: University of California, Los Angeles, CA, USA – sequence: 5 givenname: Yang orcidid: 0000-0003-3800-3533 surname: Liu fullname: Liu, Yang email: liuy03@air.tsinghua.edu.cn organization: Institute for AI Industry Research, Tsinghua University, Beijing, China |
| BookMark | eNp9kE1PAjEURRuDiYDuTdw0cT3YdtoOdUf4UBKiJGJ017zpdLQIU-wUDf_eIRAXLly9u7jnvuR0UKvylUXokpIepUTdLJ7mPUYY7aWUCiXYCWpTxWlCeCZbTSYiTUQ_ez1DnbpeEkI5V7KNxhNbzEe3eICbYANEW-CZhVC56g1PAqzttw8f-MXFdzwoYBPdl4s7HD1-8FUynY7wCCKco9MSVrW9ON4uep6MF8P7ZPZ4Nx0OZolhisYkzUxusn6ZgzQmL1U_LwVVQBTJrCgMcMOKlEsJkkoDtChKYLksjeBNpCxPu-j6sLsJ_nNr66iXfhuq5qVmQsmsTwQnTUseWib4ug621MZFiM5XMYBbaUr0XplulOm9Mn1U1oDkD7gJbg1h9x9ydUCctfa3roSQQqj0B-mid_g |
| CODEN | ITPRED |
| CitedBy_id | crossref_primary_10_1016_j_ins_2024_120956 crossref_primary_10_1109_TSP_2023_3295734 crossref_primary_10_1109_JSAC_2024_3443759 crossref_primary_10_1109_TMC_2025_3551759 crossref_primary_10_1109_TPAMI_2023_3300886 crossref_primary_10_1109_TSP_2024_3351469 crossref_primary_10_3390_electronics13173538 crossref_primary_10_1109_JIOT_2025_3546624 crossref_primary_10_1109_TSP_2023_3239799 crossref_primary_10_1007_s10115_024_02117_3 crossref_primary_10_1007_s10489_024_05956_3 crossref_primary_10_1109_TCOMM_2023_3277878 crossref_primary_10_1109_TMC_2025_3569887 crossref_primary_10_1109_TPAMI_2025_3546659 crossref_primary_10_1109_TVT_2024_3497219 crossref_primary_10_1007_s10618_022_00912_6 crossref_primary_10_1109_TSP_2023_3333658 crossref_primary_10_1109_TAI_2024_3446759 crossref_primary_10_1016_j_iot_2024_101144 crossref_primary_10_1109_TNSE_2024_3409755 crossref_primary_10_1109_TPAMI_2023_3243080 crossref_primary_10_3390_app15147843 crossref_primary_10_1016_j_neunet_2024_106772 crossref_primary_10_1109_JIOT_2023_3285937 crossref_primary_10_1038_s41598_025_99385_y crossref_primary_10_1016_j_neunet_2025_107501 crossref_primary_10_1016_j_knosys_2024_112484 crossref_primary_10_1109_TSP_2022_3168490 crossref_primary_10_1109_TSP_2024_3514802 crossref_primary_10_1109_JIOT_2024_3376552 crossref_primary_10_1109_JIOT_2024_3464239 crossref_primary_10_1109_TSP_2024_3408631 crossref_primary_10_1109_JSTSP_2024_3513692 crossref_primary_10_1109_TSP_2023_3262181 crossref_primary_10_1109_TSP_2024_3368751 crossref_primary_10_1109_TSP_2023_3241768 crossref_primary_10_3390_math12081148 crossref_primary_10_1016_j_future_2022_05_003 crossref_primary_10_3390_math12172661 crossref_primary_10_1016_j_patrec_2023_05_030 crossref_primary_10_1088_2631_8695_ad6ca3 crossref_primary_10_1109_TCE_2023_3330501 crossref_primary_10_1109_TBDATA_2024_3423693 crossref_primary_10_1016_j_procs_2025_03_151 crossref_primary_10_1109_TSP_2022_3214122 crossref_primary_10_1016_j_ifacol_2025_07_065 crossref_primary_10_1109_TII_2024_3468446 crossref_primary_10_1109_JIOT_2024_3352280 crossref_primary_10_3390_s23167235 crossref_primary_10_1109_JIOT_2024_3490855 crossref_primary_10_1109_TON_2025_3548277 crossref_primary_10_1109_TAC_2024_3383271 crossref_primary_10_1109_TNSE_2022_3185672 crossref_primary_10_1145_3718363 crossref_primary_10_1109_TSP_2023_3240083 crossref_primary_10_1109_TSP_2025_3561720 crossref_primary_10_1016_j_neunet_2024_107084 crossref_primary_10_1109_JIOT_2024_3376548 crossref_primary_10_1109_JIOT_2023_3312852 crossref_primary_10_3390_iot3020016 crossref_primary_10_1016_j_neucom_2024_127906 crossref_primary_10_1109_JIOT_2025_3534982 crossref_primary_10_1137_22M1475648 crossref_primary_10_1109_JIOT_2022_3222191 crossref_primary_10_1109_JIOT_2024_3408149 crossref_primary_10_3390_app14072720 crossref_primary_10_3390_electronics13142739 crossref_primary_10_1109_TSP_2023_3316588 crossref_primary_10_1007_s40747_023_01110_7 crossref_primary_10_1109_OJCOMS_2025_3566464 crossref_primary_10_3390_app13095270 crossref_primary_10_1109_TII_2024_3380742 crossref_primary_10_1109_ACCESS_2025_3591839 crossref_primary_10_1109_TMC_2024_3438285 crossref_primary_10_1007_s10586_024_04900_x crossref_primary_10_1109_TSP_2023_3268845 crossref_primary_10_1007_s40304_022_00332_4 crossref_primary_10_1007_s13369_025_10043_x crossref_primary_10_1016_j_comnet_2023_109657 crossref_primary_10_3390_math11081972 crossref_primary_10_1109_TSP_2023_3292497 crossref_primary_10_1109_TSP_2025_3540655 crossref_primary_10_3390_electronics14071452 crossref_primary_10_1109_TMC_2024_3510135 crossref_primary_10_1007_s10589_025_00685_w crossref_primary_10_1016_j_procs_2022_11_292 crossref_primary_10_1016_j_jfranklin_2022_12_053 crossref_primary_10_1109_OJCOMS_2024_3449563 |
| Cites_doi | 10.1007/s10463-009-0242-4 10.1109/MSP.2020.2975749 10.1109/MLSP.2016.7738878 10.1109/IEEECONF44664.2019.9049023 10.1007/s10107-019-01406-y 10.1109/TSP.2019.2943230 10.1609/aaai.v33i01.33015693 10.1007/978-1-4419-8853-9 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021 |
| DBID | 97E RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D |
| DOI | 10.1109/TSP.2021.3115952 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional |
| DatabaseTitle | CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional |
| DatabaseTitleList | Technology Research Database |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1941-0476 |
| EndPage | 6070 |
| ExternalDocumentID | 10_1109_TSP_2021_3115952 9556559 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: Air Force Office of Scientific Research; AFOSR grantid: 19RT0424 funderid: 10.13039/100000181 – fundername: ARO grantid: W911NF-19-1-0247 |
| GroupedDBID | -~X .DC 0R~ 29I 3EH 4.4 53G 5GY 5VS 6IK 85S 97E AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACIWK ACKIV ACNCT AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AJQPL AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 E.L EBS EJD F5P HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D |
| ID | FETCH-LOGICAL-c291t-37cbc78fba6ccbf98bf519a0907e5dca4c2d3466a616ca1ddfa2b6fc54ddf12b3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 110 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000717754200003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1053-587X |
| IngestDate | Mon Jun 30 10:15:23 EDT 2025 Sat Nov 29 04:10:54 EST 2025 Tue Nov 18 20:53:12 EST 2025 Wed Aug 27 02:38:52 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c291t-37cbc78fba6ccbf98bf519a0907e5dca4c2d3466a616ca1ddfa2b6fc54ddf12b3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0000-0001-6697-9731 0000-0003-3800-3533 0000-0003-1263-9365 0000-0002-1180-1415 |
| PQID | 2596780540 |
| PQPubID | 85478 |
| PageCount | 16 |
| ParticipantIDs | ieee_primary_9556559 crossref_citationtrail_10_1109_TSP_2021_3115952 crossref_primary_10_1109_TSP_2021_3115952 proquest_journals_2596780540 |
| PublicationCentury | 2000 |
| PublicationDate | 20210000 2021-00-00 20210101 |
| PublicationDateYYYYMMDD | 2021-01-01 |
| PublicationDate_xml | – year: 2021 text: 20210000 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on signal processing |
| PublicationTitleAbbrev | TSP |
| PublicationYear | 2021 |
| Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| References | wang (ref6) 2021; 22 liang (ref15) 2019 krizhevsky (ref28) 2009 bagdasaryan (ref3) 2020 caldas (ref27) 2018 urban (ref7) 2019 yu (ref8) 0 ref19 scaman (ref21) 2017 yuan (ref14) 2020; 33 zhao (ref2) 2020 cen (ref16) 2019 wei (ref4) 2020 li (ref10) 2019 sahu (ref17) 2018 ref24 ref23 ref26 ref20 ref22 (ref1) 2016 pathak (ref12) 2020 ref9 acar (ref25) 2021 khaled (ref13) 2019 ref5 karimireddy (ref11) 2020 hanzely (ref18) 2020 mcmahan (ref29) 2017 |
| References_xml | – ident: ref26 doi: 10.1007/s10463-009-0242-4 – year: 2019 ident: ref7 article-title: Local SGD converges fast and communicates little publication-title: Proc Int Conf Learn Representations – start-page: 5132 year: 2020 ident: ref11 article-title: Scaffold: Stochastic controlled averaging for on-device federated learning publication-title: Proc Int Conf Mach Learn – year: 2019 ident: ref16 article-title: Convergence of distributed stochastic variance reduced methods without sampling extra data – ident: ref9 doi: 10.1109/MSP.2020.2975749 – ident: ref23 doi: 10.1109/MLSP.2016.7738878 – year: 2019 ident: ref15 article-title: Variance reduced local SGD with lower communication complexity – volume: 22 start-page: 1 year: 2021 ident: ref6 article-title: Cooperative SGD: A unified framework for the design and analysis of communication-efficient SGD algorithms publication-title: J Mach Learn Res – year: 2020 ident: ref12 article-title: Fedsplit: An algorithmic framework for fast federated optimization – start-page: 7184 year: 0 ident: ref8 article-title: On the linear speedup analysis of communication efficient momentum SGD for distributed non-convex optimization publication-title: Proc 36th Int Conf Mach Learn – year: 2016 ident: ref1 article-title: Federated learning: Strategies for improving communication efficiency – ident: ref24 doi: 10.1109/IEEECONF44664.2019.9049023 – year: 2019 ident: ref10 article-title: On the convergence of fedavg on non-iid data publication-title: Proc Int Conf Learn Representations – volume: 33 year: 2020 ident: ref14 article-title: Federated accelerated stochastic gradient descent publication-title: Adv Neural Inf Process Syst – year: 2020 ident: ref4 article-title: A framework for evaluating gradient leakage attacks in federated learning – year: 2020 ident: ref2 article-title: iDLG: Improved Deep Leakage From Gradients – year: 2021 ident: ref25 article-title: Federated learning based on dynamic regularization publication-title: Proc Int Conf Learn Representations – ident: ref20 doi: 10.1007/s10107-019-01406-y – year: 2019 ident: ref13 article-title: First analysis of local GD on heterogeneous data – start-page: 3027 year: 2017 ident: ref21 article-title: Optimal algorithms for smooth and strongly convex distributed optimization in networks publication-title: Proc Int Conf Mach Learn – year: 2009 ident: ref28 article-title: Learning multiple layers of features from tiny images – ident: ref22 doi: 10.1109/TSP.2019.2943230 – start-page: 2938 year: 2020 ident: ref3 article-title: How to backdoor federated learning publication-title: Proc Int Conf Artif Intell Statist – ident: ref5 doi: 10.1609/aaai.v33i01.33015693 – ident: ref19 doi: 10.1007/978-1-4419-8853-9 – year: 2018 ident: ref27 article-title: Leaf: A benchmark for federated settings – year: 2020 ident: ref18 article-title: Federated learning of a mixture of global and local models – start-page: 1273 year: 2017 ident: ref29 article-title: Communication-efficient learning of deep networks from decentralized data publication-title: Proc Artif Intell Statist – year: 2018 ident: ref17 article-title: On the convergence of federated optimization in heterogeneous networks |
| SSID | ssj0014496 |
| Score | 2.687632 |
| Snippet | Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 6055 |
| SubjectTerms | Agglomeration Algorithms Communication Complexity theory Computational modeling convergence analysis data heterogeneity Data models Design optimization Distributed algorithms Distributed databases Federated learning Heterogeneity machine learning algorithms Servers Signal processing algorithms |
| Title | FedPD: A Federated Learning Framework With Adaptivity to Non-IID Data |
| URI | https://ieeexplore.ieee.org/document/9556559 https://www.proquest.com/docview/2596780540 |
| Volume | 69 |
| WOSCitedRecordID | wos000717754200003&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: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 1941-0476 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014496 issn: 1053-587X databaseCode: RIE dateStart: 19910101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH5sw4Me_DXF6ZQcvAh2a9MmbbwNZ3EgY-DU3UqaHzqQdWydf79J2w1FEbzlkJTyvTbve8l73wO4THkgPEpCB6cMO8bjEyeNBHEk15hx7VO_SPl_fgiHw2gyYaMaXG9qYZRSRfKZ6thhcZcvM7GyR2VdRgz9IKwO9TCkZa3W5sYgCIpeXIYu-A6Jwsn6StJl3fHjyASC2OtYZRlG8DcXVPRU-bERF94l3vvfe-3DbsUiUa80-wHU1OwQdr5oCzbhLlZy1L9BPRRbvQhDKSWqxFRfUbxOyUIv0_wN9SSfl10kUJ6hYTZzBoM-6vOcH8FTfDe-vXeqngmOwMzLzX4hUhFGOuVUiFSzKNWGo3HXxMCKSGEsg6UfUMqpRwX3pNQc23ofEpihh1P_GBqzbKZOABmu4XMaYhpJGjCtI2M3rKT2ZRgIV6kWdNcwJqISFLd9Ld6TIrBwWWKATyzwSQV8C642K-almMYfc5sW6M28CuMWtNeWSqq_bZmYEI7a3gyBe_r7qjPYts8uj07a0MgXK3UOW-Ijny4XF8WH9AlP5sSi |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFD7MC6gP3sV5zYMvgnVt2qSNb8NZHM4xcOreSpqLDmSTrfP3m3RtURTBtzwktHynzTknOef7AM5SHgiPktDBKcOO8fjESSNBHMk1Zlz71M9L_p86YbcbDQasV4OLqhdGKZUXn6lLO8zv8uVYzOxRWYMRE34QtgBLVjmr6Naq7gyCIFfjMgGD75AoHJSXki5r9B96JhXE3qXllmEEf3NCuarKj6049y_xxv_ebBPWizgSNeeG34KaGm3D2hd2wR24iZXsta5QE8WWMcIElRIVdKovKC6LstDzMHtFTcnf5zoSKBuj7njktNst1OIZ34XH-KZ_fesUqgmOwMzLzI4hUhFGOuVUiFSzKNUmSuOuyYIVkcLYBks_oJRTjwruSak5th0_JDBDD6f-HiyOxiO1D8hEGz6nIaaRpAHTOjKWw0pqX4aBcJWqQ6OEMREFpbhVtnhL8tTCZYkBPrHAJwXwdTivVrzP6TT-mLtjga7mFRjX4ai0VFL8b9PEJHHUqjME7sHvq05h5bZ_30k67e7dIaza58wPUo5gMZvM1DEsi49sOJ2c5B_VJ7uYx-s |
| 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=FedPD%3A+A+Federated+Learning+Framework+With+Adaptivity+to+Non-IID+Data&rft.jtitle=IEEE+transactions+on+signal+processing&rft.au=Zhang%2C+Xinwei&rft.au=Hong%2C+Mingyi&rft.au=Dhople%2C+Sairaj&rft.au=Yin%2C+Wotao&rft.date=2021&rft.pub=IEEE&rft.issn=1053-587X&rft.volume=69&rft.spage=6055&rft.epage=6070&rft_id=info:doi/10.1109%2FTSP.2021.3115952&rft.externalDocID=9556559 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1053-587X&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1053-587X&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1053-587X&client=summon |