Matches in SemOpenAlex for { <https://semopenalex.org/work/W3171987486> ?p ?o ?g. }
- W3171987486 abstract "We investigate scaling and efficiency of the deep neural network multigrid method (DNN-MG). DNN-MG is a novel neural network-based technique for the simulation of the Navier-Stokes equations that combines an adaptive geometric multigrid solver, i.e. a highly efficient classical solution scheme, with a recurrent neural network with memory. The neural network replaces in DNN-MG one or multiple finest multigrid layers and provides a correction for the classical solve in the next time step. This leads to little degradation in the solution quality while substantially reducing the overall computational costs. At the same time, the use of the multigrid solver at the coarse scales allows for a compact network that is easy to train, generalizes well, and allows for the incorporation of physical constraints. Previous work on DNN-MG focused on the overall scheme and how to enforce divergence freedom in the solution. In this work, we investigate how the network size affects training and solution quality and the overall runtime of the computations. Our results demonstrate that larger networks are able to capture the flow behavior better while requiring only little additional training time. At runtime, the use of the neural network correction can even reduce the computation time compared to a classical multigrid simulation through a faster convergence of the nonlinear solve that is required at every time step." @default.
- W3171987486 created "2021-06-22" @default.
- W3171987486 creator A5030213953 @default.
- W3171987486 creator A5030876796 @default.
- W3171987486 creator A5060011936 @default.
- W3171987486 creator A5027712423 @default.
- W3171987486 date "2021-06-14" @default.
- W3171987486 modified "2023-09-27" @default.
- W3171987486 title "Deep neural networks for geometric multigrid methods." @default.
- W3171987486 cites W1972368368 @default.
- W3171987486 cites W2026463274 @default.
- W3171987486 cites W2153944755 @default.
- W3171987486 cites W2600297185 @default.
- W3171987486 cites W2745110207 @default.
- W3171987486 cites W2899283552 @default.
- W3171987486 cites W2899771611 @default.
- W3171987486 cites W2919115771 @default.
- W3171987486 cites W2954950211 @default.
- W3171987486 cites W2963359731 @default.
- W3171987486 cites W3002852531 @default.
- W3171987486 cites W3010501116 @default.
- W3171987486 cites W3080730553 @default.
- W3171987486 cites W3092923133 @default.
- W3171987486 cites W3098011980 @default.
- W3171987486 cites W3112158738 @default.
- W3171987486 hasPublicationYear "2021" @default.
- W3171987486 type Work @default.
- W3171987486 sameAs 3171987486 @default.
- W3171987486 citedByCount "0" @default.
- W3171987486 crossrefType "posted-content" @default.
- W3171987486 hasAuthorship W3171987486A5027712423 @default.
- W3171987486 hasAuthorship W3171987486A5030213953 @default.
- W3171987486 hasAuthorship W3171987486A5030876796 @default.
- W3171987486 hasAuthorship W3171987486A5060011936 @default.
- W3171987486 hasConcept C11413529 @default.
- W3171987486 hasConcept C121332964 @default.
- W3171987486 hasConcept C126255220 @default.
- W3171987486 hasConcept C137119250 @default.
- W3171987486 hasConcept C138885662 @default.
- W3171987486 hasConcept C154945302 @default.
- W3171987486 hasConcept C158622935 @default.
- W3171987486 hasConcept C162324750 @default.
- W3171987486 hasConcept C173608175 @default.
- W3171987486 hasConcept C199360897 @default.
- W3171987486 hasConcept C207390915 @default.
- W3171987486 hasConcept C2777303404 @default.
- W3171987486 hasConcept C2778770139 @default.
- W3171987486 hasConcept C33923547 @default.
- W3171987486 hasConcept C41008148 @default.
- W3171987486 hasConcept C41895202 @default.
- W3171987486 hasConcept C45374587 @default.
- W3171987486 hasConcept C50522688 @default.
- W3171987486 hasConcept C50644808 @default.
- W3171987486 hasConcept C62520636 @default.
- W3171987486 hasConceptScore W3171987486C11413529 @default.
- W3171987486 hasConceptScore W3171987486C121332964 @default.
- W3171987486 hasConceptScore W3171987486C126255220 @default.
- W3171987486 hasConceptScore W3171987486C137119250 @default.
- W3171987486 hasConceptScore W3171987486C138885662 @default.
- W3171987486 hasConceptScore W3171987486C154945302 @default.
- W3171987486 hasConceptScore W3171987486C158622935 @default.
- W3171987486 hasConceptScore W3171987486C162324750 @default.
- W3171987486 hasConceptScore W3171987486C173608175 @default.
- W3171987486 hasConceptScore W3171987486C199360897 @default.
- W3171987486 hasConceptScore W3171987486C207390915 @default.
- W3171987486 hasConceptScore W3171987486C2777303404 @default.
- W3171987486 hasConceptScore W3171987486C2778770139 @default.
- W3171987486 hasConceptScore W3171987486C33923547 @default.
- W3171987486 hasConceptScore W3171987486C41008148 @default.
- W3171987486 hasConceptScore W3171987486C41895202 @default.
- W3171987486 hasConceptScore W3171987486C45374587 @default.
- W3171987486 hasConceptScore W3171987486C50522688 @default.
- W3171987486 hasConceptScore W3171987486C50644808 @default.
- W3171987486 hasConceptScore W3171987486C62520636 @default.
- W3171987486 hasLocation W31719874861 @default.
- W3171987486 hasOpenAccess W3171987486 @default.
- W3171987486 hasPrimaryLocation W31719874861 @default.
- W3171987486 hasRelatedWork W1929342437 @default.
- W3171987486 hasRelatedWork W2217138627 @default.
- W3171987486 hasRelatedWork W2277193644 @default.
- W3171987486 hasRelatedWork W2734734086 @default.
- W3171987486 hasRelatedWork W2898099969 @default.
- W3171987486 hasRelatedWork W2979569062 @default.
- W3171987486 hasRelatedWork W3041682155 @default.
- W3171987486 hasRelatedWork W3080114198 @default.
- W3171987486 hasRelatedWork W3093618529 @default.
- W3171987486 hasRelatedWork W3100957319 @default.
- W3171987486 hasRelatedWork W3104765509 @default.
- W3171987486 hasRelatedWork W3112158738 @default.
- W3171987486 hasRelatedWork W3131157216 @default.
- W3171987486 hasRelatedWork W3133219581 @default.
- W3171987486 hasRelatedWork W3150967811 @default.
- W3171987486 hasRelatedWork W3164380484 @default.
- W3171987486 hasRelatedWork W3170606933 @default.
- W3171987486 hasRelatedWork W3199810640 @default.
- W3171987486 hasRelatedWork W3199976834 @default.
- W3171987486 hasRelatedWork W3208781130 @default.
- W3171987486 isParatext "false" @default.
- W3171987486 isRetracted "false" @default.
- W3171987486 magId "3171987486" @default.