Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210101648> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W3210101648 abstract "Federated learning allows edge devices to learn a shared global model from the client’s model parameters while keeping the training data on the device. However, for large models, transmitting all model parameters imposes considerable communication costs, which can be a significant bottleneck in bandwidth-constrained deployments. We present a federated matched averaging algorithm with information-gain based sampling that considerably reduces the number of parameters to be sent by all clients in a federated learning paradigm. Experiments across five benchmark datasets, encompassing symbolic, image and audio data, suggest that our algorithm significantly reduces communication overhead in federated learning settings, without reducing classification accuracy." @default.
- W3210101648 created "2021-11-08" @default.
- W3210101648 creator A5075281480 @default.
- W3210101648 creator A5081335569 @default.
- W3210101648 date "2021-10-21" @default.
- W3210101648 modified "2023-09-26" @default.
- W3210101648 title "Federated matched averaging with information-gain based parameter sampling" @default.
- W3210101648 cites W2108475251 @default.
- W3210101648 cites W2222512263 @default.
- W3210101648 cites W2810470036 @default.
- W3210101648 cites W2912654452 @default.
- W3210101648 cites W3103802018 @default.
- W3210101648 cites W4206320562 @default.
- W3210101648 doi "https://doi.org/10.1145/3486001.3486225" @default.
- W3210101648 hasPublicationYear "2021" @default.
- W3210101648 type Work @default.
- W3210101648 sameAs 3210101648 @default.
- W3210101648 citedByCount "3" @default.
- W3210101648 countsByYear W32101016482022 @default.
- W3210101648 countsByYear W32101016482023 @default.
- W3210101648 crossrefType "proceedings-article" @default.
- W3210101648 hasAuthorship W3210101648A5075281480 @default.
- W3210101648 hasAuthorship W3210101648A5081335569 @default.
- W3210101648 hasConcept C111919701 @default.
- W3210101648 hasConcept C119857082 @default.
- W3210101648 hasConcept C124101348 @default.
- W3210101648 hasConcept C13280743 @default.
- W3210101648 hasConcept C149635348 @default.
- W3210101648 hasConcept C154945302 @default.
- W3210101648 hasConcept C162307627 @default.
- W3210101648 hasConcept C185798385 @default.
- W3210101648 hasConcept C205649164 @default.
- W3210101648 hasConcept C2776257435 @default.
- W3210101648 hasConcept C2779960059 @default.
- W3210101648 hasConcept C2780513914 @default.
- W3210101648 hasConcept C2992525071 @default.
- W3210101648 hasConcept C31258907 @default.
- W3210101648 hasConcept C41008148 @default.
- W3210101648 hasConceptScore W3210101648C111919701 @default.
- W3210101648 hasConceptScore W3210101648C119857082 @default.
- W3210101648 hasConceptScore W3210101648C124101348 @default.
- W3210101648 hasConceptScore W3210101648C13280743 @default.
- W3210101648 hasConceptScore W3210101648C149635348 @default.
- W3210101648 hasConceptScore W3210101648C154945302 @default.
- W3210101648 hasConceptScore W3210101648C162307627 @default.
- W3210101648 hasConceptScore W3210101648C185798385 @default.
- W3210101648 hasConceptScore W3210101648C205649164 @default.
- W3210101648 hasConceptScore W3210101648C2776257435 @default.
- W3210101648 hasConceptScore W3210101648C2779960059 @default.
- W3210101648 hasConceptScore W3210101648C2780513914 @default.
- W3210101648 hasConceptScore W3210101648C2992525071 @default.
- W3210101648 hasConceptScore W3210101648C31258907 @default.
- W3210101648 hasConceptScore W3210101648C41008148 @default.
- W3210101648 hasLocation W32101016481 @default.
- W3210101648 hasOpenAccess W3210101648 @default.
- W3210101648 hasPrimaryLocation W32101016481 @default.
- W3210101648 hasRelatedWork W2157013742 @default.
- W3210101648 hasRelatedWork W2998565617 @default.
- W3210101648 hasRelatedWork W3036485160 @default.
- W3210101648 hasRelatedWork W3189674571 @default.
- W3210101648 hasRelatedWork W4221143749 @default.
- W3210101648 hasRelatedWork W4225517547 @default.
- W3210101648 hasRelatedWork W4226502461 @default.
- W3210101648 hasRelatedWork W4309969787 @default.
- W3210101648 hasRelatedWork W4317464145 @default.
- W3210101648 hasRelatedWork W4386100056 @default.
- W3210101648 isParatext "false" @default.
- W3210101648 isRetracted "false" @default.
- W3210101648 magId "3210101648" @default.
- W3210101648 workType "article" @default.