Matches in SemOpenAlex for { <https://semopenalex.org/work/W4311034768> ?p ?o ?g. }
- W4311034768 abstract "Modeling three-dimensional (3D) turbulence by neural networks is difficult because 3D turbulence is highly nonlinear with high degrees of freedom and the corresponding simulation is memory-intensive. Recently, the attention mechanism has been shown as a promising approach to boost the performance of neural networks on turbulence simulation. However, the standard self-attention mechanism uses O(n2) time and space with respect to input dimension n, and such quadratic complexity has become the main bottleneck for attention to be applied on 3D turbulence simulation. In this work, we resolve this issue with the concept of a linear attention network. The linear attention approximates the standard attention by adding two linear projections, reducing the overall self-attention complexity from O(n2) to O(n) in both time and space. The linear attention coupled Fourier neural operator (LAFNO) is developed for the simulation of 3D isotropic turbulence and free shear turbulence. Numerical simulations show that the linear attention mechanism provides 40% error reduction at the same level of computational cost, and LAFNO can accurately reconstruct a variety of statistics and instantaneous spatial structures of 3D turbulence. The linear attention method would be helpful for the improvement of neural network models of 3D nonlinear problems involving high-dimensional data in other scientific domains." @default.
- W4311034768 created "2022-12-22" @default.
- W4311034768 creator A5020850738 @default.
- W4311034768 creator A5037862100 @default.
- W4311034768 creator A5062274120 @default.
- W4311034768 creator A5078654535 @default.
- W4311034768 date "2023-01-01" @default.
- W4311034768 modified "2023-09-30" @default.
- W4311034768 title "Linear attention coupled Fourier neural operator for simulation of three-dimensional turbulence" @default.
- W4311034768 cites W1501987263 @default.
- W4311034768 cites W2149298154 @default.
- W4311034768 cites W2534240011 @default.
- W4311034768 cites W2749028154 @default.
- W4311034768 cites W2777417212 @default.
- W4311034768 cites W2795982117 @default.
- W4311034768 cites W2807826281 @default.
- W4311034768 cites W2896618593 @default.
- W4311034768 cites W2899283552 @default.
- W4311034768 cites W2919115771 @default.
- W4311034768 cites W2920959147 @default.
- W4311034768 cites W2946771678 @default.
- W4311034768 cites W2948567396 @default.
- W4311034768 cites W2980396542 @default.
- W4311034768 cites W2990629279 @default.
- W4311034768 cites W3010839048 @default.
- W4311034768 cites W3025593802 @default.
- W4311034768 cites W3080891795 @default.
- W4311034768 cites W3093325407 @default.
- W4311034768 cites W3097844861 @default.
- W4311034768 cites W3098633050 @default.
- W4311034768 cites W3102140816 @default.
- W4311034768 cites W3103841771 @default.
- W4311034768 cites W3105245152 @default.
- W4311034768 cites W3112714421 @default.
- W4311034768 cites W3124389259 @default.
- W4311034768 cites W3126393581 @default.
- W4311034768 cites W3129768217 @default.
- W4311034768 cites W3167256391 @default.
- W4311034768 cites W3180745686 @default.
- W4311034768 cites W3181235980 @default.
- W4311034768 cites W3193983211 @default.
- W4311034768 cites W3197308578 @default.
- W4311034768 cites W3198740648 @default.
- W4311034768 cites W3201253676 @default.
- W4311034768 cites W3203127985 @default.
- W4311034768 cites W4200610765 @default.
- W4311034768 cites W4223521822 @default.
- W4311034768 cites W4226082378 @default.
- W4311034768 cites W4285604295 @default.
- W4311034768 cites W4288734997 @default.
- W4311034768 cites W4291002711 @default.
- W4311034768 cites W4293764686 @default.
- W4311034768 cites W4296380561 @default.
- W4311034768 cites W4310497802 @default.
- W4311034768 doi "https://doi.org/10.1063/5.0130334" @default.
- W4311034768 hasPublicationYear "2023" @default.
- W4311034768 type Work @default.
- W4311034768 citedByCount "4" @default.
- W4311034768 countsByYear W43110347682023 @default.
- W4311034768 crossrefType "journal-article" @default.
- W4311034768 hasAuthorship W4311034768A5020850738 @default.
- W4311034768 hasAuthorship W4311034768A5037862100 @default.
- W4311034768 hasAuthorship W4311034768A5062274120 @default.
- W4311034768 hasAuthorship W4311034768A5078654535 @default.
- W4311034768 hasBestOaLocation W43110347682 @default.
- W4311034768 hasConcept C11413529 @default.
- W4311034768 hasConcept C114148568 @default.
- W4311034768 hasConcept C120665830 @default.
- W4311034768 hasConcept C121332964 @default.
- W4311034768 hasConcept C121864883 @default.
- W4311034768 hasConcept C149635348 @default.
- W4311034768 hasConcept C150711758 @default.
- W4311034768 hasConcept C154945302 @default.
- W4311034768 hasConcept C158622935 @default.
- W4311034768 hasConcept C168110828 @default.
- W4311034768 hasConcept C184050105 @default.
- W4311034768 hasConcept C196558001 @default.
- W4311034768 hasConcept C202444582 @default.
- W4311034768 hasConcept C2780513914 @default.
- W4311034768 hasConcept C33676613 @default.
- W4311034768 hasConcept C33923547 @default.
- W4311034768 hasConcept C41008148 @default.
- W4311034768 hasConcept C50644808 @default.
- W4311034768 hasConcept C57879066 @default.
- W4311034768 hasConcept C62520636 @default.
- W4311034768 hasConcept C76155785 @default.
- W4311034768 hasConceptScore W4311034768C11413529 @default.
- W4311034768 hasConceptScore W4311034768C114148568 @default.
- W4311034768 hasConceptScore W4311034768C120665830 @default.
- W4311034768 hasConceptScore W4311034768C121332964 @default.
- W4311034768 hasConceptScore W4311034768C121864883 @default.
- W4311034768 hasConceptScore W4311034768C149635348 @default.
- W4311034768 hasConceptScore W4311034768C150711758 @default.
- W4311034768 hasConceptScore W4311034768C154945302 @default.
- W4311034768 hasConceptScore W4311034768C158622935 @default.
- W4311034768 hasConceptScore W4311034768C168110828 @default.
- W4311034768 hasConceptScore W4311034768C184050105 @default.
- W4311034768 hasConceptScore W4311034768C196558001 @default.
- W4311034768 hasConceptScore W4311034768C202444582 @default.
- W4311034768 hasConceptScore W4311034768C2780513914 @default.