Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319654190> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W4319654190 endingPage "928" @default.
- W4319654190 startingPage "915" @default.
- W4319654190 abstract "Distributed Stein Variational Gradient Descent (DSVGD) is a non-parametric distributed learning framework for federated Bayesian learning, where multiple clients jointly train a machine learning model by communicating a number of non-random and interacting particles with the server. Since communication resources are limited, selecting the clients with most informative local learning updates can improve the model convergence and communication efficiency. In this paper, we propose two selection schemes for DSVGD based on Kernelized Stein Discrepancy (KSD) and Hilbert Inner Product (HIP). We derive the upper bound on the decrease of the global free energy per iteration for both schemes, which is then minimized to speed up the model convergence. We evaluate and compare our schemes with conventional schemes in terms of model accuracy, convergence speed, and stability using various learning tasks and datasets." @default.
- W4319654190 created "2023-02-10" @default.
- W4319654190 creator A5019105341 @default.
- W4319654190 creator A5046214153 @default.
- W4319654190 creator A5060023455 @default.
- W4319654190 date "2023-04-01" @default.
- W4319654190 modified "2023-10-12" @default.
- W4319654190 title "Client Selection for Federated Bayesian Learning" @default.
- W4319654190 cites W1590183771 @default.
- W4319654190 cites W1901616594 @default.
- W4319654190 cites W2912213068 @default.
- W4319654190 cites W2920095265 @default.
- W4319654190 cites W2963318081 @default.
- W4319654190 cites W2963932270 @default.
- W4319654190 cites W2978015420 @default.
- W4319654190 cites W2981138228 @default.
- W4319654190 cites W2999074226 @default.
- W4319654190 cites W3009934096 @default.
- W4319654190 cites W3021378343 @default.
- W4319654190 cites W3027223193 @default.
- W4319654190 cites W3033843856 @default.
- W4319654190 cites W3059999910 @default.
- W4319654190 cites W3084967198 @default.
- W4319654190 cites W3103802018 @default.
- W4319654190 cites W3108590825 @default.
- W4319654190 cites W3110962653 @default.
- W4319654190 cites W3113075536 @default.
- W4319654190 cites W3120614079 @default.
- W4319654190 cites W4205197926 @default.
- W4319654190 doi "https://doi.org/10.1109/jsac.2023.3242720" @default.
- W4319654190 hasPublicationYear "2023" @default.
- W4319654190 type Work @default.
- W4319654190 citedByCount "0" @default.
- W4319654190 crossrefType "journal-article" @default.
- W4319654190 hasAuthorship W4319654190A5019105341 @default.
- W4319654190 hasAuthorship W4319654190A5046214153 @default.
- W4319654190 hasAuthorship W4319654190A5060023455 @default.
- W4319654190 hasBestOaLocation W43196541902 @default.
- W4319654190 hasConcept C105795698 @default.
- W4319654190 hasConcept C107673813 @default.
- W4319654190 hasConcept C112972136 @default.
- W4319654190 hasConcept C117251300 @default.
- W4319654190 hasConcept C119857082 @default.
- W4319654190 hasConcept C134306372 @default.
- W4319654190 hasConcept C154945302 @default.
- W4319654190 hasConcept C15744967 @default.
- W4319654190 hasConcept C162324750 @default.
- W4319654190 hasConcept C19417346 @default.
- W4319654190 hasConcept C206688291 @default.
- W4319654190 hasConcept C2777303404 @default.
- W4319654190 hasConcept C2779582901 @default.
- W4319654190 hasConcept C33923547 @default.
- W4319654190 hasConcept C41008148 @default.
- W4319654190 hasConcept C50522688 @default.
- W4319654190 hasConcept C50644808 @default.
- W4319654190 hasConcept C77553402 @default.
- W4319654190 hasConcept C81917197 @default.
- W4319654190 hasConceptScore W4319654190C105795698 @default.
- W4319654190 hasConceptScore W4319654190C107673813 @default.
- W4319654190 hasConceptScore W4319654190C112972136 @default.
- W4319654190 hasConceptScore W4319654190C117251300 @default.
- W4319654190 hasConceptScore W4319654190C119857082 @default.
- W4319654190 hasConceptScore W4319654190C134306372 @default.
- W4319654190 hasConceptScore W4319654190C154945302 @default.
- W4319654190 hasConceptScore W4319654190C15744967 @default.
- W4319654190 hasConceptScore W4319654190C162324750 @default.
- W4319654190 hasConceptScore W4319654190C19417346 @default.
- W4319654190 hasConceptScore W4319654190C206688291 @default.
- W4319654190 hasConceptScore W4319654190C2777303404 @default.
- W4319654190 hasConceptScore W4319654190C2779582901 @default.
- W4319654190 hasConceptScore W4319654190C33923547 @default.
- W4319654190 hasConceptScore W4319654190C41008148 @default.
- W4319654190 hasConceptScore W4319654190C50522688 @default.
- W4319654190 hasConceptScore W4319654190C50644808 @default.
- W4319654190 hasConceptScore W4319654190C77553402 @default.
- W4319654190 hasConceptScore W4319654190C81917197 @default.
- W4319654190 hasFunder F4320321001 @default.
- W4319654190 hasIssue "4" @default.
- W4319654190 hasLocation W43196541901 @default.
- W4319654190 hasLocation W43196541902 @default.
- W4319654190 hasOpenAccess W4319654190 @default.
- W4319654190 hasPrimaryLocation W43196541901 @default.
- W4319654190 hasRelatedWork W2790129348 @default.
- W4319654190 hasRelatedWork W2792987183 @default.
- W4319654190 hasRelatedWork W2961085424 @default.
- W4319654190 hasRelatedWork W2989932438 @default.
- W4319654190 hasRelatedWork W3012534984 @default.
- W4319654190 hasRelatedWork W4220975826 @default.
- W4319654190 hasRelatedWork W4307415224 @default.
- W4319654190 hasRelatedWork W4311414921 @default.
- W4319654190 hasRelatedWork W4312393190 @default.
- W4319654190 hasRelatedWork W4315777593 @default.
- W4319654190 hasVolume "41" @default.
- W4319654190 isParatext "false" @default.
- W4319654190 isRetracted "false" @default.
- W4319654190 workType "article" @default.