Self-Balancing Federated Learning With Global Imbalanced Data in Mobile Systems
Federated learning (FL) is a distributed deep learning method that enables multiple participants, such as mobile and IoT devices, to contribute a neural network while their private training data remains in local devices. This distributed approach is promising in the mobile systems where have a large...
Saved in:
| Published in: | IEEE transactions on parallel and distributed systems Vol. 32; no. 1; pp. 59 - 71 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
New York
IEEE
01.01.2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects: | |
| ISSN: | 1045-9219, 1558-2183 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Federated learning (FL) is a distributed deep learning method that enables multiple participants, such as mobile and IoT devices, to contribute a neural network while their private training data remains in local devices. This distributed approach is promising in the mobile systems where have a large corpus of decentralized data and require high privacy. However, unlike the common datasets, the data distribution of the mobile systems is imbalanced which will increase the bias of model. In this article, we demonstrate that the imbalanced distributed training data will cause an accuracy degradation of FL applications. To counter this problem, we build a self-balancing FL framework named Astraea, which alleviates the imbalances by 1) Z-score-based data augmentation, and 2) Mediator-based multi-client rescheduling. The proposed framework relieves global imbalance by adaptive data augmentation and downsampling, and for averaging the local imbalance, it creates the mediator to reschedule the training of clients based on Kullback-Leibler divergence (KLD) of their data distribution. Compared with FedAvg , the vanilla FL algorithm, Astraea shows +4.39 and +6.51 percent improvement of top-1 accuracy on the imbalanced EMNIST and imbalanced CINIC-10 datasets, respectively. Meanwhile, the communication traffic of Astraea is reduced by 75 percent compared to FedAvg . |
|---|---|
| AbstractList | Federated learning (FL) is a distributed deep learning method that enables multiple participants, such as mobile and IoT devices, to contribute a neural network while their private training data remains in local devices. This distributed approach is promising in the mobile systems where have a large corpus of decentralized data and require high privacy. However, unlike the common datasets, the data distribution of the mobile systems is imbalanced which will increase the bias of model. In this article, we demonstrate that the imbalanced distributed training data will cause an accuracy degradation of FL applications. To counter this problem, we build a self-balancing FL framework named Astraea, which alleviates the imbalances by 1) Z-score-based data augmentation, and 2) Mediator-based multi-client rescheduling. The proposed framework relieves global imbalance by adaptive data augmentation and downsampling, and for averaging the local imbalance, it creates the mediator to reschedule the training of clients based on Kullback-Leibler divergence (KLD) of their data distribution. Compared with FedAvg , the vanilla FL algorithm, Astraea shows +4.39 and +6.51 percent improvement of top-1 accuracy on the imbalanced EMNIST and imbalanced CINIC-10 datasets, respectively. Meanwhile, the communication traffic of Astraea is reduced by 75 percent compared to FedAvg . |
| Author | Liang, Liang Liu, Duo Duan, Moming Chen, Xianzhang Tan, Yujuan Liu, Renping |
| Author_xml | – sequence: 1 givenname: Moming surname: Duan fullname: Duan, Moming email: booker_duan@yahoo.com organization: Key Laboratory of Dependable Service Computing in Cyber Physical Society, the Ministry of Education, Chongqing University, Chongqing, China – sequence: 2 givenname: Duo orcidid: 0000-0002-3040-2065 surname: Liu fullname: Liu, Duo email: liuduo@cqu.edu.cn organization: Key Laboratory of Dependable Service Computing in Cyber Physical Society, the Ministry of Education, Chongqing University, Chongqing, China – sequence: 3 givenname: Xianzhang surname: Chen fullname: Chen, Xianzhang email: xzchen@cqu.edu.cn organization: Key Laboratory of Dependable Service Computing in Cyber Physical Society, the Ministry of Education, Chongqing University, Chongqing, China – sequence: 4 givenname: Renping surname: Liu fullname: Liu, Renping email: liurenping123321@163.com organization: Key Laboratory of Dependable Service Computing in Cyber Physical Society, the Ministry of Education, Chongqing University, Chongqing, China – sequence: 5 givenname: Yujuan orcidid: 0000-0002-9055-5389 surname: Tan fullname: Tan, Yujuan email: tanyujuan@gmail.com organization: Key Laboratory of Dependable Service Computing in Cyber Physical Society, the Ministry of Education, Chongqing University, Chongqing, China – sequence: 6 givenname: Liang orcidid: 0000-0002-2778-455X surname: Liang fullname: Liang, Liang email: liangliang@cqu.edu.cn organization: School of Microelectronics and Communication Engineering, Chongqing University, Chongqing, China |
| BookMark | eNp9kM1PwkAQxTcGExH9A4yXJp6Ls1_9OCoIkmAwAeNxs9tOdUlpcXc58N9bhHjw4GleZt6bl_wuSa9pGyTkhsKQUsjvV6_j5ZABgyEHyAUkZ6RPpcxiRjPe6zQIGeeM5hfk0vs1ABUSRJ8sllhX8aOudVPY5iOaYIlOByyjOWrXHFbvNnxG07o1uo5mG_Nj7e5jHXRkm-ilNbbGaLn3ATf-ipxXuvZ4fZoD8jZ5Wo2e4_liOhs9zOOC5TzEZZ5LzNBkkCa61JJLSQWUpWGJMUyklcwYNVkBqAvkElIpTFlVFZSdSmTFB-Tu-Hfr2q8d-qDW7c41XaVigtMUhBBZ56JHV-Fa7x1WauvsRru9oqAO3NSBmzpwUyduXSb9kyls0MG2TXDa1v8mb49Ji4i_TTkVVPCEfwObJnvO |
| CODEN | ITDSEO |
| CitedBy_id | crossref_primary_10_1016_j_knosys_2023_110813 crossref_primary_10_1109_TIFS_2024_3516547 crossref_primary_10_1016_j_jnca_2025_104105 crossref_primary_10_1109_TBDATA_2022_3231277 crossref_primary_10_1109_TPDS_2023_3334398 crossref_primary_10_1109_TKDE_2023_3341808 crossref_primary_10_1109_TSUSC_2024_3403427 crossref_primary_10_2298_CSIS250131052Z crossref_primary_10_3390_s24082579 crossref_primary_10_1016_j_ins_2022_12_045 crossref_primary_10_1186_s40537_025_01195_6 crossref_primary_10_1007_s12083_021_01254_8 crossref_primary_10_1109_TMC_2022_3147792 crossref_primary_10_1109_TSC_2023_3332102 crossref_primary_10_3390_s22020450 crossref_primary_10_1016_j_asoc_2024_111483 crossref_primary_10_1109_JIOT_2023_3299947 crossref_primary_10_1109_JIOT_2022_3153343 crossref_primary_10_1109_TC_2023_3333213 crossref_primary_10_1007_s44196_025_00814_7 crossref_primary_10_1109_LWC_2022_3147236 crossref_primary_10_1109_TCE_2023_3321846 crossref_primary_10_1109_TMC_2024_3398052 crossref_primary_10_1109_TSMC_2024_3422068 crossref_primary_10_1109_TASE_2023_3270354 crossref_primary_10_1109_TSC_2025_3536309 crossref_primary_10_1007_s11390_024_3639_x crossref_primary_10_1051_wujns_2022276508 crossref_primary_10_1109_TNNLS_2024_3417452 crossref_primary_10_1109_TIFS_2022_3227435 crossref_primary_10_1109_JIOT_2024_3370964 crossref_primary_10_1016_j_compeleceng_2024_110012 crossref_primary_10_1109_TNET_2024_3363916 crossref_primary_10_1145_3730587 crossref_primary_10_3390_electronics14122331 crossref_primary_10_1109_JIOT_2024_3502245 crossref_primary_10_26599_BDMA_2024_9020067 crossref_primary_10_1109_TPWRS_2023_3267463 crossref_primary_10_1109_COMST_2024_3427324 crossref_primary_10_1016_j_cose_2023_103597 crossref_primary_10_1016_j_conengprac_2024_106098 crossref_primary_10_1109_JIOT_2023_3271692 crossref_primary_10_1109_TIFS_2022_3186739 crossref_primary_10_1109_TPDS_2022_3150579 crossref_primary_10_1016_j_future_2024_04_026 crossref_primary_10_1109_TMC_2023_3315451 crossref_primary_10_1016_j_oceaneng_2025_120823 crossref_primary_10_1016_j_cosrev_2024_100697 crossref_primary_10_1016_j_future_2022_07_010 crossref_primary_10_1016_j_jksuci_2024_101912 crossref_primary_10_1145_3623397 crossref_primary_10_1016_j_comnet_2025_111406 crossref_primary_10_1016_j_ins_2024_120475 crossref_primary_10_1016_j_iot_2022_100642 crossref_primary_10_1007_s40747_023_01332_9 crossref_primary_10_1109_ACCESS_2023_3267964 crossref_primary_10_1109_COMST_2024_3352910 crossref_primary_10_1109_TNSE_2024_3410275 crossref_primary_10_1016_j_future_2024_107626 crossref_primary_10_1016_j_jbi_2024_104712 crossref_primary_10_1109_TPDS_2022_3151853 crossref_primary_10_1109_TSUSC_2023_3341836 crossref_primary_10_1109_ACCESS_2024_3392838 crossref_primary_10_1109_TMC_2023_3262829 crossref_primary_10_1007_s10586_025_05260_w crossref_primary_10_1109_TAFFC_2024_3516822 crossref_primary_10_1109_ACCESS_2024_3413069 crossref_primary_10_1016_j_asoc_2021_107872 crossref_primary_10_1109_TNSE_2021_3053588 crossref_primary_10_1007_s10844_023_00797_x crossref_primary_10_1109_TMC_2024_3388731 crossref_primary_10_1109_TKDE_2024_3372708 crossref_primary_10_1109_TNSM_2025_3560629 crossref_primary_10_1109_JIOT_2021_3130115 crossref_primary_10_1109_TVT_2022_3205307 crossref_primary_10_1109_TCE_2024_3355058 crossref_primary_10_1109_TPDS_2021_3134263 crossref_primary_10_1109_JIOT_2023_3341811 crossref_primary_10_1016_j_compind_2021_103509 crossref_primary_10_1109_JIOT_2023_3286398 crossref_primary_10_1016_j_comnet_2025_111428 crossref_primary_10_1145_3606017 crossref_primary_10_1109_TPDS_2021_3098467 crossref_primary_10_3390_fi17070315 crossref_primary_10_1016_j_future_2021_10_016 crossref_primary_10_1109_TPAMI_2023_3290213 crossref_primary_10_1007_s10489_022_04070_6 crossref_primary_10_1007_s11042_023_17737_0 crossref_primary_10_1007_s10586_025_05250_y crossref_primary_10_1109_TSC_2025_3528325 crossref_primary_10_1016_j_future_2021_09_015 crossref_primary_10_1109_ACCESS_2024_3353617 crossref_primary_10_1109_TCOMM_2023_3302067 crossref_primary_10_1186_s42400_024_00295_9 crossref_primary_10_3390_electronics13214193 crossref_primary_10_1109_TMI_2023_3299206 crossref_primary_10_1145_3429252 crossref_primary_10_1016_j_neunet_2024_106436 crossref_primary_10_1109_TNSE_2022_3168969 crossref_primary_10_1109_COMST_2024_3399612 crossref_primary_10_1109_TAI_2023_3254519 crossref_primary_10_1109_TNNLS_2022_3160699 crossref_primary_10_1016_j_eswa_2025_128600 crossref_primary_10_1109_TMC_2024_3452312 crossref_primary_10_1109_JIOT_2021_3089054 crossref_primary_10_1016_j_future_2025_107911 crossref_primary_10_1016_j_hcc_2025_100325 crossref_primary_10_1016_j_ins_2024_120825 crossref_primary_10_1109_TPDS_2021_3127712 crossref_primary_10_1007_s11063_024_11493_4 crossref_primary_10_1109_TSC_2024_3387734 crossref_primary_10_1007_s10586_025_05163_w crossref_primary_10_1109_ACCESS_2022_3156045 crossref_primary_10_1109_TMC_2025_3539284 crossref_primary_10_1109_TMC_2023_3310248 crossref_primary_10_1109_COMST_2022_3200740 crossref_primary_10_1109_TPDS_2023_3240767 crossref_primary_10_1109_TPDS_2022_3218807 crossref_primary_10_1016_j_eswa_2023_121982 crossref_primary_10_1016_j_knosys_2024_111420 crossref_primary_10_1109_TWC_2022_3226867 crossref_primary_10_1016_j_conengprac_2024_106050 crossref_primary_10_1109_TPDS_2021_3139191 crossref_primary_10_1016_j_knosys_2023_110658 crossref_primary_10_1109_JSYST_2023_3274197 crossref_primary_10_1109_TSC_2021_3109910 crossref_primary_10_1109_TPDS_2023_3244198 crossref_primary_10_1016_j_future_2025_108114 crossref_primary_10_1016_j_dt_2024_08_015 crossref_primary_10_1109_TMC_2024_3466227 crossref_primary_10_1016_j_patcog_2025_111746 crossref_primary_10_1007_s11704_023_2791_8 crossref_primary_10_1109_TNSE_2022_3196463 crossref_primary_10_1016_j_iot_2023_100890 crossref_primary_10_1016_j_knosys_2025_113494 crossref_primary_10_1093_jcde_qwaf068 crossref_primary_10_3390_e27060553 crossref_primary_10_1016_j_inffus_2025_103555 crossref_primary_10_1109_TITS_2024_3362058 crossref_primary_10_1109_TMC_2025_3528404 crossref_primary_10_1145_3708495 crossref_primary_10_1016_j_engappai_2025_111201 crossref_primary_10_1109_TSC_2022_3187962 crossref_primary_10_1016_j_inffus_2023_101923 crossref_primary_10_1109_TNNLS_2025_3552206 crossref_primary_10_1109_TMC_2025_3558874 crossref_primary_10_1109_TMC_2024_3509852 crossref_primary_10_1109_TAI_2021_3133819 crossref_primary_10_1109_ACCESS_2024_3435910 crossref_primary_10_3390_sym17050782 crossref_primary_10_1109_JIOT_2024_3358799 crossref_primary_10_1007_s00530_025_01719_3 crossref_primary_10_1016_j_neucom_2024_127906 crossref_primary_10_1145_3659205 crossref_primary_10_1038_s41598_025_11864_4 crossref_primary_10_1007_s00221_022_06430_7 crossref_primary_10_1109_ACCESS_2022_3184309 crossref_primary_10_1109_JIOT_2024_3467110 crossref_primary_10_1109_TMC_2022_3223578 crossref_primary_10_1109_TKDE_2024_3523043 crossref_primary_10_1016_j_knosys_2023_110763 crossref_primary_10_1016_j_comnet_2024_110663 crossref_primary_10_1109_TPDS_2023_3238768 crossref_primary_10_1109_ACCESS_2022_3142899 crossref_primary_10_1109_TWC_2022_3162595 crossref_primary_10_1109_TPDS_2022_3142314 crossref_primary_10_1109_TII_2022_3149516 crossref_primary_10_1007_s11704_024_40065_x crossref_primary_10_1109_TC_2024_3474180 crossref_primary_10_1016_j_future_2020_12_003 crossref_primary_10_1109_JIOT_2023_3262582 crossref_primary_10_1016_j_eswa_2023_122874 crossref_primary_10_1109_TPDS_2023_3264480 crossref_primary_10_1109_JIOT_2023_3301019 crossref_primary_10_1016_j_eswa_2024_123460 crossref_primary_10_1109_TCE_2024_3478349 crossref_primary_10_1109_TMC_2024_3502686 crossref_primary_10_3390_s24010088 crossref_primary_10_1007_s10115_024_02285_2 crossref_primary_10_1007_s10844_024_00910_8 crossref_primary_10_1016_j_future_2025_108037 crossref_primary_10_1109_TII_2022_3190034 crossref_primary_10_1109_TPDS_2024_3501581 crossref_primary_10_1145_3674973 crossref_primary_10_1007_s11227_025_07374_9 crossref_primary_10_1016_j_jiixd_2023_08_003 crossref_primary_10_1016_j_patcog_2023_109507 crossref_primary_10_1109_ACCESS_2024_3508030 crossref_primary_10_3390_e25040606 crossref_primary_10_1109_ACCESS_2024_3351600 crossref_primary_10_1109_TPDS_2022_3225185 crossref_primary_10_1109_TPDS_2020_3040981 crossref_primary_10_1007_s10462_023_10563_8 crossref_primary_10_1016_j_comcom_2023_04_023 crossref_primary_10_1109_TC_2021_3099723 crossref_primary_10_1016_j_neunet_2022_07_030 crossref_primary_10_1016_j_future_2024_107595 crossref_primary_10_1109_TNSE_2024_3447904 crossref_primary_10_1109_TSG_2023_3318418 crossref_primary_10_1109_TSC_2023_3332703 crossref_primary_10_1109_TITS_2022_3149753 crossref_primary_10_1109_TNSE_2024_3483295 crossref_primary_10_1145_3735125 crossref_primary_10_1109_TVT_2023_3277438 crossref_primary_10_3390_e25050715 crossref_primary_10_1016_j_eswa_2024_125336 crossref_primary_10_1109_LWC_2025_3556462 crossref_primary_10_1016_j_comcom_2023_04_012 crossref_primary_10_1109_TMC_2024_3521304 crossref_primary_10_1007_s10044_025_01533_7 crossref_primary_10_1109_TMC_2022_3200016 crossref_primary_10_1016_j_knosys_2025_113073 crossref_primary_10_1007_s12083_024_01660_8 crossref_primary_10_1007_s10462_022_10141_4 crossref_primary_10_1016_j_neucom_2022_10_012 crossref_primary_10_1109_TPDS_2024_3480115 crossref_primary_10_1007_s11432_024_4254_5 crossref_primary_10_1049_blc2_12054 crossref_primary_10_1109_ACCESS_2020_3032520 crossref_primary_10_1109_TPDS_2021_3134647 crossref_primary_10_1016_j_hcc_2022_100068 crossref_primary_10_3390_fi14110338 |
| Cites_doi | 10.1109/TKDE.2008.239 10.1109/CVPR.2018.00568 10.1109/MSP.2012.2205597 10.1613/jair.953 10.1145/2939672.2939785 10.14778/2212351.2212354 10.1109/IJCNN.2017.7966217 10.1038/nature14539 10.1016/j.future.2018.09.020 10.1109/ICCD46524.2019.00038 10.1109/DICTA.2016.7797091 10.1007/978-3-319-57959-7 10.1016/j.ijmedinf.2018.01.007 10.1145/3065386 10.1145/3133956.3133982 10.1109/INFOCOM.2019.8737416 10.1145/3298981 10.1109/TNNLS.2019.2944481 10.1109/GLOCOM.2018.8647927 10.1023/A:1007618119488 10.1007/s00365-006-0663-2 10.1109/CVPR.2019.00482 10.1109/TIFS.2019.2929409 |
| 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/TPDS.2020.3009406 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present 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 Computer Science |
| EISSN | 1558-2183 |
| EndPage | 71 |
| ExternalDocumentID | 10_1109_TPDS_2020_3009406 9141436 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: National Natural Science Foundation of China grantid: 61672116; 61601067; 61802038; 61672115 funderid: 10.13039/501100001809 – fundername: Chongqing Youth Talent Support Program – fundername: Fundamental Research Funds for the Central Universities grantid: 0214005207005; 2019CDJGFJSJ001 funderid: 10.13039/501100012226 – fundername: Chongqing High-Tech Research Key Program grantid: cstc2019jscx-mbdx0063 – fundername: China Postdoctoral Science Foundation grantid: 2017M620412 funderid: 10.13039/501100002858 |
| GroupedDBID | --Z -~X .DC 0R~ 29I 4.4 5GY 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACIWK AENEX AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD HZ~ IEDLZ IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS TN5 TWZ UHB AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D |
| ID | FETCH-LOGICAL-c293t-d995e8eb8076ada5355140ddb26bb247f5821b8c0eace350754bdfff0d75465f3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 325 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000557346500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1045-9219 |
| IngestDate | Sun Nov 09 06:00:21 EST 2025 Sat Nov 29 06:06:47 EST 2025 Tue Nov 18 22:26:35 EST 2025 Wed Aug 27 02:31:22 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| 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-c293t-d995e8eb8076ada5355140ddb26bb247f5821b8c0eace350754bdfff0d75465f3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0000-0002-9055-5389 0000-0002-3040-2065 0000-0002-2778-455X |
| PQID | 2431704448 |
| PQPubID | 85437 |
| PageCount | 13 |
| ParticipantIDs | crossref_primary_10_1109_TPDS_2020_3009406 crossref_citationtrail_10_1109_TPDS_2020_3009406 proquest_journals_2431704448 ieee_primary_9141436 |
| PublicationCentury | 2000 |
| PublicationDate | 2021-Jan.-1 2021-1-1 20210101 |
| PublicationDateYYYYMMDD | 2021-01-01 |
| PublicationDate_xml | – month: 01 year: 2021 text: 2021-Jan.-1 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on parallel and distributed systems |
| PublicationTitleAbbrev | TPDS |
| 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 | ref13 ref12 smith (ref25) 2017 luo (ref49) 2019 lecun (ref2) 2015; 521 ref17 chen (ref48) 2018 ref18 abadi (ref20) 2016 adamic (ref39) 2002; 3 zhao (ref15) 2018 ref46 yang (ref45) 2018 lecun (ref14) 2010; 2 han (ref36) 2005 ref4 ref3 finn (ref47) 2017 hard (ref10) 2018 ref6 ref5 mohri (ref33) 2019 dean (ref21) 2012 ref35 ref34 ref37 bonawitz (ref11) 2019 ref31 ref30 ref32 ref1 mcmahan (ref9) 2017 ref38 wan (ref42) 2013; 28 darlow (ref19) 2018 agarwal (ref29) 2018 zhang (ref8) 2015 ref24 ref26 kingma (ref43) 2014 ref22 ref28 ref27 li (ref7) 2014 krizhevsky (ref16) 2009 iglewicz (ref44) 1993; 16 bodaghi (ref41) 2017 hard (ref23) 2018 caldas (ref40) 2018 |
| References_xml | – ident: ref34 doi: 10.1109/TKDE.2008.239 – start-page: 685 year: 2015 ident: ref8 article-title: Deep learning with elastic averaging SGD publication-title: Proc 28th Int Conf Neural Inf Process Syst – ident: ref3 doi: 10.1109/CVPR.2018.00568 – volume: 3 start-page: 143 year: 2002 ident: ref39 article-title: Zipf's law and the internet publication-title: Glottometrics – year: 2018 ident: ref23 article-title: Federated learning for mobile keyboard prediction – start-page: 1126 year: 2017 ident: ref47 article-title: Model-agnostic meta-learning for fast adaptation of deep networks publication-title: Proc 34th Int Conf Mach Learn – ident: ref5 doi: 10.1109/MSP.2012.2205597 – ident: ref35 doi: 10.1613/jair.953 – ident: ref38 doi: 10.1145/2939672.2939785 – year: 2019 ident: ref49 article-title: Real-world image datasets for federated learning – ident: ref22 doi: 10.14778/2212351.2212354 – ident: ref17 doi: 10.1109/IJCNN.2017.7966217 – volume: 28 start-page: 1058 year: 2013 ident: ref42 article-title: Regularization of neural networks using dropconnect publication-title: Proc 30th Int Conf Mach Learn – start-page: 4427 year: 2017 ident: ref25 article-title: Federated multi-task learning publication-title: Proc 31st Int Conf Neural Inf Process Syst – volume: 521 year: 2015 ident: ref2 article-title: Deep learning publication-title: Nature doi: 10.1038/nature14539 – year: 2018 ident: ref40 article-title: Leaf: A benchmark for federated settings – ident: ref27 doi: 10.1016/j.future.2018.09.020 – year: 2017 ident: ref41 article-title: Dynamics of instagram users – ident: ref1 doi: 10.1109/ICCD46524.2019.00038 – year: 2009 ident: ref16 article-title: Learning multiple layers of features from tiny images – ident: ref18 doi: 10.1109/DICTA.2016.7797091 – ident: ref6 doi: 10.1007/978-3-319-57959-7 – ident: ref24 doi: 10.1016/j.ijmedinf.2018.01.007 – year: 2019 ident: ref11 article-title: Towards federated learning at scale: System design publication-title: Proc 2nd SysML Conf – ident: ref13 doi: 10.1145/3065386 – year: 2014 ident: ref43 article-title: Adam: A method for stochastic optimization – year: 2018 ident: ref10 article-title: Federated learning for mobile keyboard prediction – ident: ref28 doi: 10.1145/3133956.3133982 – year: 2018 ident: ref45 article-title: Applied federated learning: Improving google keyboard query suggestions – ident: ref30 doi: 10.1109/INFOCOM.2019.8737416 – start-page: 583 year: 2014 ident: ref7 article-title: Scaling distributed machine learning with the parameter server publication-title: Proc 11th USENIX Symp Operating Syst Des Implementation – year: 2018 ident: ref15 article-title: Federated learning with non-IID data – start-page: 4615 year: 2019 ident: ref33 article-title: Agnostic federated learning publication-title: Proceedings 36th Int Conf Mach Learn – ident: ref12 doi: 10.1145/3298981 – year: 2018 ident: ref48 article-title: Federated meta-learning for recommendation – ident: ref32 doi: 10.1109/TNNLS.2019.2944481 – start-page: 265 year: 2016 ident: ref20 article-title: Tensorflow: A system for large-scale machine learning publication-title: Proc 12th USENIX Symp Operating Syst Des Implementation – start-page: 7564 year: 2018 ident: ref29 article-title: cpSGD: Communication-efficient and differentially-private distributed SGD publication-title: Proc 32nd Int Conf Neural Inf Process Syst – start-page: 1273 year: 2017 ident: ref9 article-title: Communication-efficient learning of deep networks from decentralized data publication-title: Proc 20th Int Conf Artif Intell Statist – ident: ref26 doi: 10.1109/GLOCOM.2018.8647927 – volume: 2 year: 2010 ident: ref14 article-title: MNIST handwritten digit database publication-title: ATT Labs – ident: ref37 doi: 10.1023/A:1007618119488 – year: 2018 ident: ref19 article-title: CINIC-10 is not imagenet or CIFAR-10 – ident: ref46 doi: 10.1007/s00365-006-0663-2 – ident: ref4 doi: 10.1109/CVPR.2019.00482 – start-page: 878 year: 2005 ident: ref36 article-title: Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning publication-title: Proc Int Conf Intell Comput – start-page: 1223 year: 2012 ident: ref21 article-title: Large scale distributed deep networks publication-title: Proc 25th Int Conf Neural Inf Process Syst – volume: 16 year: 1993 ident: ref44 publication-title: How to Detect and Handle Outliers – ident: ref31 doi: 10.1109/TIFS.2019.2929409 |
| SSID | ssj0014504 |
| Score | 2.7029996 |
| Snippet | Federated learning (FL) is a distributed deep learning method that enables multiple participants, such as mobile and IoT devices, to contribute a neural... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 59 |
| SubjectTerms | Algorithms Balancing Communications traffic Data augmentation Data models Datasets Distributed databases distributed machine learning Electronic devices Federated learning Machine learning Mobile communication systems Mobile handsets Neural networks Privacy Rescheduling Servers Training |
| Title | Self-Balancing Federated Learning With Global Imbalanced Data in Mobile Systems |
| URI | https://ieeexplore.ieee.org/document/9141436 https://www.proquest.com/docview/2431704448 |
| Volume | 32 |
| WOSCitedRecordID | wos000557346500001&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: 1558-2183 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014504 issn: 1045-9219 databaseCode: RIE dateStart: 19900101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEJ0g8aAHUdCIounBk3GldD-6PapI9CCSgJHbpl-rJAgGFn-_bSmERGPibbM7TTb7tp2ZzvQ9gIuE8VQIwgLsyoycs4DJOAo4TxQWsXPiTmyCdrvpcMh6Jbhan4XRWrvmM31tL10tX03lwm6VNVkrMu492YItSunyrNa6YhDFTirQZBdxwMw09BXMFmbNQa_dN5kgMQmqbaSz4kYbPsiJqvxYiZ176VT-92L7sOfDSHSzxP0ASnpShcpKogH5GVuF3Q2-wRo89_U4D25tN6M0N1DHMkmYYFMhT7P6hl5HxTtaCgGgxw_hTM3zNi84Gk3Q01SYZQR5nvNDeOncD-4eAq-oEEjj1otAMRbrVIsU04QrHoc2XsJKCZIYwCKa22OzIpXYLMc6NKFiHAmV5zlW1Iqm5-ERlCfTiT4G1FIkpJFOiUxYFBIhZMpkyCNpApA8pqQOePWNM-npxq3qxThzaQdmmYUls7BkHpY6XK6HfC65Nv4yrlkc1oYegjo0VkBmfjbOM2KjJEuMl578PuoUdojtVXFbKw0oF7OFPoNt-VWM5rNz96N9AwAuzhA |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFD7oFNQH7-J0ah58EuuyNGmbR29jQ52CE30ruVUHuomr_n6TLBsDRfCttCdQ-jU55-ScfB_AYcJFJiXhEfZlRiF4xBWjkRCJxpJ5J-7FJtJOJ3t64nczcDw5C2OM8c1n5sRd-lq-HqhPt1VW5w1q3XsyC3OMUtIYndaa1Awo82KBNr9gEbcTMdQwG5jXu3cX9zYXJDZFda10Tt5oygt5WZUfa7F3MM2V_73aKiyHQBKdjpBfgxnTX4eVsUgDCnN2HZamGAc34PbevBbRmetnVPYGajouCRtuahSIVp_RY698QSMpANR-k97UPr8QpUC9ProZSLuQoMB0vgkPzcvueSsKmgqRso69jDTnzGRGZjhNhBYsdhET1lqSxEJG08IdnJWZwnZBNrENFhmVuigKrFMnm17EW1DpD_pmG1BDkzilJiMq4TQmUqqMq1hQZUOQgqWkCnj8jXMVCMed7sVr7hMPzHMHS-5gyQMsVTiaDHkfsW38ZbzhcJgYBgiqUBsDmYf5OMyJi5McNV628_uoA1hodW-u8-t252oXFonrXPEbLTWolB-fZg_m1VfZG37s-5_uG-NT0Vc |
| 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=Self-Balancing+Federated+Learning+With+Global+Imbalanced+Data+in+Mobile+Systems&rft.jtitle=IEEE+transactions+on+parallel+and+distributed+systems&rft.au=Duan%2C+Moming&rft.au=Liu%2C+Duo&rft.au=Chen%2C+Xianzhang&rft.au=Liu%2C+Renping&rft.date=2021-01-01&rft.issn=1045-9219&rft.eissn=1558-2183&rft.volume=32&rft.issue=1&rft.spage=59&rft.epage=71&rft_id=info:doi/10.1109%2FTPDS.2020.3009406&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TPDS_2020_3009406 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1045-9219&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1045-9219&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1045-9219&client=summon |