Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384502087> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4384502087 endingPage "403" @default.
- W4384502087 startingPage "392" @default.
- W4384502087 abstract "Understanding and optimizing the parallel training strategies in the training of large-scale Deep Neural Network (DNN) models is crucial to enhance training efficiency. Existing works tried to demonstrate the layer-level information of the computational graph to support parallel training strategy selection. Whereas, the overall parallel execution logic is rarely considered by previous methods. In this paper, we proposed a novel visual analytics approach for parallel training strategies, demonstrating the execution logic of the distributed computing from up to bottom via explaining communication operators. Specifically, a computation-communication bipartite construction algorithm is designed for the computational graph visualization. Furthermore, a system is developed to help users easily access the proposed approach and explore the parallel training strategies interactively. With empirical evaluation through a quantitative user study and a qualitative expert interview, the practicality and superiority of the proposed approach is verified." @default.
- W4384502087 created "2023-07-18" @default.
- W4384502087 creator A5010441587 @default.
- W4384502087 creator A5011981325 @default.
- W4384502087 creator A5013438405 @default.
- W4384502087 creator A5017541508 @default.
- W4384502087 creator A5045499782 @default.
- W4384502087 creator A5045702383 @default.
- W4384502087 creator A5049718129 @default.
- W4384502087 creator A5064049630 @default.
- W4384502087 creator A5066730504 @default.
- W4384502087 date "2023-10-01" @default.
- W4384502087 modified "2023-10-01" @default.
- W4384502087 title "Interactive visual analytics of parallel training strategies for DNN models" @default.
- W4384502087 cites W2004049427 @default.
- W4384502087 cites W2049084163 @default.
- W4384502087 cites W2052963087 @default.
- W4384502087 cites W2071353749 @default.
- W4384502087 cites W2126536832 @default.
- W4384502087 cites W2133068391 @default.
- W4384502087 cites W2135415614 @default.
- W4384502087 cites W2154172361 @default.
- W4384502087 cites W2194775991 @default.
- W4384502087 cites W2475334473 @default.
- W4384502087 cites W2607223307 @default.
- W4384502087 cites W2618530766 @default.
- W4384502087 cites W2752332392 @default.
- W4384502087 cites W2753713840 @default.
- W4384502087 cites W2957783787 @default.
- W4384502087 cites W2969388332 @default.
- W4384502087 cites W3007205676 @default.
- W4384502087 cites W3047537431 @default.
- W4384502087 cites W3125301056 @default.
- W4384502087 cites W3205746072 @default.
- W4384502087 cites W3209029431 @default.
- W4384502087 cites W4213060883 @default.
- W4384502087 cites W4281866107 @default.
- W4384502087 cites W4286947131 @default.
- W4384502087 doi "https://doi.org/10.1016/j.cag.2023.07.030" @default.
- W4384502087 hasPublicationYear "2023" @default.
- W4384502087 type Work @default.
- W4384502087 citedByCount "0" @default.
- W4384502087 crossrefType "journal-article" @default.
- W4384502087 hasAuthorship W4384502087A5010441587 @default.
- W4384502087 hasAuthorship W4384502087A5011981325 @default.
- W4384502087 hasAuthorship W4384502087A5013438405 @default.
- W4384502087 hasAuthorship W4384502087A5017541508 @default.
- W4384502087 hasAuthorship W4384502087A5045499782 @default.
- W4384502087 hasAuthorship W4384502087A5045702383 @default.
- W4384502087 hasAuthorship W4384502087A5049718129 @default.
- W4384502087 hasAuthorship W4384502087A5064049630 @default.
- W4384502087 hasAuthorship W4384502087A5066730504 @default.
- W4384502087 hasConcept C11413529 @default.
- W4384502087 hasConcept C119857082 @default.
- W4384502087 hasConcept C132525143 @default.
- W4384502087 hasConcept C154945302 @default.
- W4384502087 hasConcept C197657726 @default.
- W4384502087 hasConcept C36464697 @default.
- W4384502087 hasConcept C41008148 @default.
- W4384502087 hasConcept C45374587 @default.
- W4384502087 hasConcept C50644808 @default.
- W4384502087 hasConcept C59732488 @default.
- W4384502087 hasConcept C80444323 @default.
- W4384502087 hasConceptScore W4384502087C11413529 @default.
- W4384502087 hasConceptScore W4384502087C119857082 @default.
- W4384502087 hasConceptScore W4384502087C132525143 @default.
- W4384502087 hasConceptScore W4384502087C154945302 @default.
- W4384502087 hasConceptScore W4384502087C197657726 @default.
- W4384502087 hasConceptScore W4384502087C36464697 @default.
- W4384502087 hasConceptScore W4384502087C41008148 @default.
- W4384502087 hasConceptScore W4384502087C45374587 @default.
- W4384502087 hasConceptScore W4384502087C50644808 @default.
- W4384502087 hasConceptScore W4384502087C59732488 @default.
- W4384502087 hasConceptScore W4384502087C80444323 @default.
- W4384502087 hasLocation W43845020871 @default.
- W4384502087 hasOpenAccess W4384502087 @default.
- W4384502087 hasPrimaryLocation W43845020871 @default.
- W4384502087 hasRelatedWork W1568173680 @default.
- W4384502087 hasRelatedWork W2090914728 @default.
- W4384502087 hasRelatedWork W2119485335 @default.
- W4384502087 hasRelatedWork W2362367986 @default.
- W4384502087 hasRelatedWork W2377311228 @default.
- W4384502087 hasRelatedWork W2961085424 @default.
- W4384502087 hasRelatedWork W3208185614 @default.
- W4384502087 hasRelatedWork W4236163602 @default.
- W4384502087 hasRelatedWork W4240492454 @default.
- W4384502087 hasRelatedWork W4306674287 @default.
- W4384502087 hasVolume "115" @default.
- W4384502087 isParatext "false" @default.
- W4384502087 isRetracted "false" @default.
- W4384502087 workType "article" @default.