Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313287653> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W4313287653 endingPage "4788" @default.
- W4313287653 startingPage "4788" @default.
- W4313287653 abstract "AI provides a new method for massive simulated data calculations in molecular dynamics, materials, and other scientific computing fields. However, the complex structures and large-scale parameters of neural network models make them difficult to develop and train. The automatic parallel technology based on graph algorithms is one of the most promising methods to solve this problem, despite the low efficiency in the design, implementation, and execution of distributed parallel policies for large-scale neural network models. In this paper, we propose an adaptive distributed parallel training method based on the dynamic generation of critical DAG (directed acyclic graph) paths, called FD-DPS, to solve this efficiency problem. Firstly, the proposed model splits operators with the dimension of the tensor, which can expand the space available for model parallelism. Secondly, a dynamic critical path generation method is employed to determine node priority changes in the DAG of the neural network models. Finally, the model implements the optimal scheduling of critical paths based on the priority of the nodes, thereby improving the performance of parallel strategies. Our experiments show that FD-DPS can achieve 12.76% and 11.78% faster training on PnasNet_mobile and ResNet_200 models, respectively, compared with the MP-DPS and Fast methods." @default.
- W4313287653 created "2023-01-06" @default.
- W4313287653 creator A5008416156 @default.
- W4313287653 creator A5012416453 @default.
- W4313287653 creator A5046597133 @default.
- W4313287653 creator A5064680757 @default.
- W4313287653 creator A5074556294 @default.
- W4313287653 creator A5089731861 @default.
- W4313287653 date "2022-12-16" @default.
- W4313287653 modified "2023-09-24" @default.
- W4313287653 title "Adaptive Distributed Parallel Training Method for a Deep Learning Model Based on Dynamic Critical Paths of DAG" @default.
- W4313287653 cites W1591439598 @default.
- W4313287653 cites W1981893977 @default.
- W4313287653 cites W2113358325 @default.
- W4313287653 cites W2630499550 @default.
- W4313287653 cites W2883830791 @default.
- W4313287653 cites W2897761084 @default.
- W4313287653 cites W2902572901 @default.
- W4313287653 cites W2950130535 @default.
- W4313287653 cites W2968923792 @default.
- W4313287653 cites W2985776732 @default.
- W4313287653 cites W3011495430 @default.
- W4313287653 cites W3019389960 @default.
- W4313287653 cites W3080400811 @default.
- W4313287653 cites W3096858590 @default.
- W4313287653 cites W3100064097 @default.
- W4313287653 cites W3112483307 @default.
- W4313287653 cites W3177828909 @default.
- W4313287653 cites W3201073812 @default.
- W4313287653 cites W4206174365 @default.
- W4313287653 cites W4293230379 @default.
- W4313287653 doi "https://doi.org/10.3390/math10244788" @default.
- W4313287653 hasPublicationYear "2022" @default.
- W4313287653 type Work @default.
- W4313287653 citedByCount "0" @default.
- W4313287653 crossrefType "journal-article" @default.
- W4313287653 hasAuthorship W4313287653A5008416156 @default.
- W4313287653 hasAuthorship W4313287653A5012416453 @default.
- W4313287653 hasAuthorship W4313287653A5046597133 @default.
- W4313287653 hasAuthorship W4313287653A5064680757 @default.
- W4313287653 hasAuthorship W4313287653A5074556294 @default.
- W4313287653 hasAuthorship W4313287653A5089731861 @default.
- W4313287653 hasBestOaLocation W43132876531 @default.
- W4313287653 hasConcept C11413529 @default.
- W4313287653 hasConcept C115874739 @default.
- W4313287653 hasConcept C120314980 @default.
- W4313287653 hasConcept C120373497 @default.
- W4313287653 hasConcept C126255220 @default.
- W4313287653 hasConcept C127413603 @default.
- W4313287653 hasConcept C154945302 @default.
- W4313287653 hasConcept C173608175 @default.
- W4313287653 hasConcept C199360897 @default.
- W4313287653 hasConcept C201995342 @default.
- W4313287653 hasConcept C206729178 @default.
- W4313287653 hasConcept C26713055 @default.
- W4313287653 hasConcept C33923547 @default.
- W4313287653 hasConcept C41008148 @default.
- W4313287653 hasConcept C50644808 @default.
- W4313287653 hasConcept C74197172 @default.
- W4313287653 hasConcept C80444323 @default.
- W4313287653 hasConceptScore W4313287653C11413529 @default.
- W4313287653 hasConceptScore W4313287653C115874739 @default.
- W4313287653 hasConceptScore W4313287653C120314980 @default.
- W4313287653 hasConceptScore W4313287653C120373497 @default.
- W4313287653 hasConceptScore W4313287653C126255220 @default.
- W4313287653 hasConceptScore W4313287653C127413603 @default.
- W4313287653 hasConceptScore W4313287653C154945302 @default.
- W4313287653 hasConceptScore W4313287653C173608175 @default.
- W4313287653 hasConceptScore W4313287653C199360897 @default.
- W4313287653 hasConceptScore W4313287653C201995342 @default.
- W4313287653 hasConceptScore W4313287653C206729178 @default.
- W4313287653 hasConceptScore W4313287653C26713055 @default.
- W4313287653 hasConceptScore W4313287653C33923547 @default.
- W4313287653 hasConceptScore W4313287653C41008148 @default.
- W4313287653 hasConceptScore W4313287653C50644808 @default.
- W4313287653 hasConceptScore W4313287653C74197172 @default.
- W4313287653 hasConceptScore W4313287653C80444323 @default.
- W4313287653 hasIssue "24" @default.
- W4313287653 hasLocation W43132876531 @default.
- W4313287653 hasLocation W43132876532 @default.
- W4313287653 hasOpenAccess W4313287653 @default.
- W4313287653 hasPrimaryLocation W43132876531 @default.
- W4313287653 hasRelatedWork W156781440 @default.
- W4313287653 hasRelatedWork W1591171990 @default.
- W4313287653 hasRelatedWork W1882733036 @default.
- W4313287653 hasRelatedWork W1992741870 @default.
- W4313287653 hasRelatedWork W2100117254 @default.
- W4313287653 hasRelatedWork W2117014006 @default.
- W4313287653 hasRelatedWork W2160425906 @default.
- W4313287653 hasRelatedWork W2546696010 @default.
- W4313287653 hasRelatedWork W2997221951 @default.
- W4313287653 hasRelatedWork W3019093293 @default.
- W4313287653 hasVolume "10" @default.
- W4313287653 isParatext "false" @default.
- W4313287653 isRetracted "false" @default.
- W4313287653 workType "article" @default.