Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206207913> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W3206207913 endingPage "467" @default.
- W3206207913 startingPage "459" @default.
- W3206207913 abstract "Previous chapter Next chapter Full AccessProceedings Proceedings of the 2022 SIAM International Conference on Data Mining (SDM)FedMe: Federated Learning via Model ExchangeKoji Matsuda, Yuya Sasaki, Chuan Xiao, and Makoto OnizukaKoji Matsuda, Yuya Sasaki, Chuan Xiao, and Makoto Onizukapp.459 - 467Chapter DOI:https://doi.org/10.1137/1.9781611977172.52PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract Federated learning is a distributed machine learning method in which a single server and multiple clients collaboratively build machine learning models without sharing datasets on clients. Numerous methods have been proposed to cope with the data heterogeneity issue in federated learning. Existing solutions require a model architecture tuned by the central server, yet a major technical challenge is that it is difficult to tune the model architecture due to the absence of local data on the central server. In this paper, we propose Federated learning via Model exchange (FedMe), which personalizes models with automatic model architecture tuning during the learning process. The novelty of FedMe lies in its learning process: clients exchange their models for model architecture tuning and model training. First, to optimize the model architectures for local data, clients tune their own personalized models by comparing to exchanged models and picking the one that yields the best performance. Second, clients train both personalized models and exchanged models by using deep mutual learning, in spite of different model architectures across the clients. We perform experiments on three real datasets and show that FedMe outperforms state-of-the-art federated learning methods while tuning model architectures automatically. Previous chapter Next chapter RelatedDetails Published:2022eISBN:978-1-61197-717-2 https://doi.org/10.1137/1.9781611977172Book Series Name:ProceedingsBook Code:PRDT22Book Pages:1-737" @default.
- W3206207913 created "2021-10-25" @default.
- W3206207913 creator A5030272842 @default.
- W3206207913 creator A5032790180 @default.
- W3206207913 creator A5036148682 @default.
- W3206207913 creator A5041703913 @default.
- W3206207913 date "2022-01-01" @default.
- W3206207913 modified "2023-10-16" @default.
- W3206207913 title "FedMe: Federated Learning via Model Exchange" @default.
- W3206207913 cites W134960717 @default.
- W3206207913 cites W2127218421 @default.
- W3206207913 cites W2541884796 @default.
- W3206207913 cites W2620998106 @default.
- W3206207913 cites W2769956639 @default.
- W3206207913 cites W2896422817 @default.
- W3206207913 cites W2900120080 @default.
- W3206207913 cites W2902113386 @default.
- W3206207913 cites W2917462349 @default.
- W3206207913 cites W2945996667 @default.
- W3206207913 cites W2963374479 @default.
- W3206207913 cites W2970971581 @default.
- W3206207913 cites W2980216952 @default.
- W3206207913 cites W2982312326 @default.
- W3206207913 cites W2995022099 @default.
- W3206207913 cites W2995653155 @default.
- W3206207913 cites W2996736801 @default.
- W3206207913 cites W3007548213 @default.
- W3206207913 cites W3008187686 @default.
- W3206207913 cites W3015636663 @default.
- W3206207913 cites W3017371741 @default.
- W3206207913 cites W3034906194 @default.
- W3206207913 cites W3035668299 @default.
- W3206207913 cites W3038022836 @default.
- W3206207913 cites W3039612675 @default.
- W3206207913 cites W3091870957 @default.
- W3206207913 cites W3100393648 @default.
- W3206207913 cites W3118608800 @default.
- W3206207913 cites W3122967774 @default.
- W3206207913 cites W3125494587 @default.
- W3206207913 doi "https://doi.org/10.1137/1.9781611977172.52" @default.
- W3206207913 hasPublicationYear "2022" @default.
- W3206207913 type Work @default.
- W3206207913 sameAs 3206207913 @default.
- W3206207913 citedByCount "5" @default.
- W3206207913 countsByYear W32062079132022 @default.
- W3206207913 countsByYear W32062079132023 @default.
- W3206207913 crossrefType "book-chapter" @default.
- W3206207913 hasAuthorship W3206207913A5030272842 @default.
- W3206207913 hasAuthorship W3206207913A5032790180 @default.
- W3206207913 hasAuthorship W3206207913A5036148682 @default.
- W3206207913 hasAuthorship W3206207913A5041703913 @default.
- W3206207913 hasBestOaLocation W32062079132 @default.
- W3206207913 hasConcept C108583219 @default.
- W3206207913 hasConcept C111919701 @default.
- W3206207913 hasConcept C119857082 @default.
- W3206207913 hasConcept C123657996 @default.
- W3206207913 hasConcept C138885662 @default.
- W3206207913 hasConcept C142362112 @default.
- W3206207913 hasConcept C153349607 @default.
- W3206207913 hasConcept C154945302 @default.
- W3206207913 hasConcept C27206212 @default.
- W3206207913 hasConcept C2778738651 @default.
- W3206207913 hasConcept C2992525071 @default.
- W3206207913 hasConcept C41008148 @default.
- W3206207913 hasConcept C98045186 @default.
- W3206207913 hasConceptScore W3206207913C108583219 @default.
- W3206207913 hasConceptScore W3206207913C111919701 @default.
- W3206207913 hasConceptScore W3206207913C119857082 @default.
- W3206207913 hasConceptScore W3206207913C123657996 @default.
- W3206207913 hasConceptScore W3206207913C138885662 @default.
- W3206207913 hasConceptScore W3206207913C142362112 @default.
- W3206207913 hasConceptScore W3206207913C153349607 @default.
- W3206207913 hasConceptScore W3206207913C154945302 @default.
- W3206207913 hasConceptScore W3206207913C27206212 @default.
- W3206207913 hasConceptScore W3206207913C2778738651 @default.
- W3206207913 hasConceptScore W3206207913C2992525071 @default.
- W3206207913 hasConceptScore W3206207913C41008148 @default.
- W3206207913 hasConceptScore W3206207913C98045186 @default.
- W3206207913 hasLocation W32062079131 @default.
- W3206207913 hasLocation W32062079132 @default.
- W3206207913 hasOpenAccess W3206207913 @default.
- W3206207913 hasPrimaryLocation W32062079131 @default.
- W3206207913 hasRelatedWork W3014300295 @default.
- W3206207913 hasRelatedWork W3164822677 @default.
- W3206207913 hasRelatedWork W4223943233 @default.
- W3206207913 hasRelatedWork W4225161397 @default.
- W3206207913 hasRelatedWork W4250304930 @default.
- W3206207913 hasRelatedWork W4312200629 @default.
- W3206207913 hasRelatedWork W4360585206 @default.
- W3206207913 hasRelatedWork W4364306694 @default.
- W3206207913 hasRelatedWork W4380075502 @default.
- W3206207913 hasRelatedWork W4380086463 @default.
- W3206207913 isParatext "false" @default.
- W3206207913 isRetracted "false" @default.
- W3206207913 magId "3206207913" @default.
- W3206207913 workType "book-chapter" @default.