Matches in SemOpenAlex for { <https://semopenalex.org/work/W4283211323> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W4283211323 abstract "Federated learning enables different parties to collaboratively build a global model under the orchestration of a server while keeping the training data on clients' devices. However, performance is affected when clients have heterogeneous data. To cope with this problem, we assume that despite data heterogeneity, there are groups of clients who have similar data distributions that can be clustered. In previous approaches, in order to cluster clients the server requires clients to send their parameters simultaneously. However, this can be problematic in a context where there is a significant number of participants that may have limited availability. To prevent such a bottleneck, we propose FLIC (Federated Learning with Incremental Clustering), in which the server exploits the updates sent by clients during federated training instead of asking them to send their parameters simultaneously. Hence no additional communications between the server and the clients are necessary other than what classical federated learning requires. We empirically demonstrate for various non-IID cases that our approach successfully splits clients into groups following the same data distributions. We also identify the limitations of FLIC by studying its capability to partition clients at the early stages of the federated learning process efficiently. We further address attacks on models as a form of data heterogeneity and empirically show that FLIC is a robust defense against poisoning attacks even when the proportion of malicious clients is higher than 50%." @default.
- W4283211323 created "2022-06-22" @default.
- W4283211323 creator A5004199718 @default.
- W4283211323 creator A5059969352 @default.
- W4283211323 creator A5072736415 @default.
- W4283211323 creator A5079237780 @default.
- W4283211323 date "2022-06-17" @default.
- W4283211323 modified "2023-09-27" @default.
- W4283211323 title "Federated learning with incremental clustering for heterogeneous data" @default.
- W4283211323 doi "https://doi.org/10.48550/arxiv.2206.08752" @default.
- W4283211323 hasPublicationYear "2022" @default.
- W4283211323 type Work @default.
- W4283211323 citedByCount "0" @default.
- W4283211323 crossrefType "posted-content" @default.
- W4283211323 hasAuthorship W4283211323A5004199718 @default.
- W4283211323 hasAuthorship W4283211323A5059969352 @default.
- W4283211323 hasAuthorship W4283211323A5072736415 @default.
- W4283211323 hasAuthorship W4283211323A5079237780 @default.
- W4283211323 hasBestOaLocation W42832113231 @default.
- W4283211323 hasConcept C10138342 @default.
- W4283211323 hasConcept C111919701 @default.
- W4283211323 hasConcept C114614502 @default.
- W4283211323 hasConcept C119857082 @default.
- W4283211323 hasConcept C120314980 @default.
- W4283211323 hasConcept C124101348 @default.
- W4283211323 hasConcept C142362112 @default.
- W4283211323 hasConcept C149635348 @default.
- W4283211323 hasConcept C151730666 @default.
- W4283211323 hasConcept C153349607 @default.
- W4283211323 hasConcept C162324750 @default.
- W4283211323 hasConcept C164866538 @default.
- W4283211323 hasConcept C165696696 @default.
- W4283211323 hasConcept C182306322 @default.
- W4283211323 hasConcept C199168358 @default.
- W4283211323 hasConcept C2779343474 @default.
- W4283211323 hasConcept C2780513914 @default.
- W4283211323 hasConcept C2992525071 @default.
- W4283211323 hasConcept C31258907 @default.
- W4283211323 hasConcept C33923547 @default.
- W4283211323 hasConcept C38652104 @default.
- W4283211323 hasConcept C41008148 @default.
- W4283211323 hasConcept C42812 @default.
- W4283211323 hasConcept C558565934 @default.
- W4283211323 hasConcept C73555534 @default.
- W4283211323 hasConcept C86803240 @default.
- W4283211323 hasConcept C98045186 @default.
- W4283211323 hasConceptScore W4283211323C10138342 @default.
- W4283211323 hasConceptScore W4283211323C111919701 @default.
- W4283211323 hasConceptScore W4283211323C114614502 @default.
- W4283211323 hasConceptScore W4283211323C119857082 @default.
- W4283211323 hasConceptScore W4283211323C120314980 @default.
- W4283211323 hasConceptScore W4283211323C124101348 @default.
- W4283211323 hasConceptScore W4283211323C142362112 @default.
- W4283211323 hasConceptScore W4283211323C149635348 @default.
- W4283211323 hasConceptScore W4283211323C151730666 @default.
- W4283211323 hasConceptScore W4283211323C153349607 @default.
- W4283211323 hasConceptScore W4283211323C162324750 @default.
- W4283211323 hasConceptScore W4283211323C164866538 @default.
- W4283211323 hasConceptScore W4283211323C165696696 @default.
- W4283211323 hasConceptScore W4283211323C182306322 @default.
- W4283211323 hasConceptScore W4283211323C199168358 @default.
- W4283211323 hasConceptScore W4283211323C2779343474 @default.
- W4283211323 hasConceptScore W4283211323C2780513914 @default.
- W4283211323 hasConceptScore W4283211323C2992525071 @default.
- W4283211323 hasConceptScore W4283211323C31258907 @default.
- W4283211323 hasConceptScore W4283211323C33923547 @default.
- W4283211323 hasConceptScore W4283211323C38652104 @default.
- W4283211323 hasConceptScore W4283211323C41008148 @default.
- W4283211323 hasConceptScore W4283211323C42812 @default.
- W4283211323 hasConceptScore W4283211323C558565934 @default.
- W4283211323 hasConceptScore W4283211323C73555534 @default.
- W4283211323 hasConceptScore W4283211323C86803240 @default.
- W4283211323 hasConceptScore W4283211323C98045186 @default.
- W4283211323 hasLocation W42832113231 @default.
- W4283211323 hasLocation W42832113232 @default.
- W4283211323 hasOpenAccess W4283211323 @default.
- W4283211323 hasPrimaryLocation W42832113231 @default.
- W4283211323 hasRelatedWork W2092479881 @default.
- W4283211323 hasRelatedWork W2208347154 @default.
- W4283211323 hasRelatedWork W2356889451 @default.
- W4283211323 hasRelatedWork W2380798983 @default.
- W4283211323 hasRelatedWork W2751317954 @default.
- W4283211323 hasRelatedWork W2888523397 @default.
- W4283211323 hasRelatedWork W3010198599 @default.
- W4283211323 hasRelatedWork W4283211323 @default.
- W4283211323 hasRelatedWork W4312970728 @default.
- W4283211323 hasRelatedWork W1544908136 @default.
- W4283211323 isParatext "false" @default.
- W4283211323 isRetracted "false" @default.
- W4283211323 workType "article" @default.