Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204279182> ?p ?o ?g. }
- W3204279182 abstract "This paper addresses variational data assimilation from a learning point of view. Data assimilation aims to reconstruct the time evolution of some state given a series of observations, possibly noisy and irregularly-sampled. Using automatic differentiation tools embedded in deep learning frameworks, we introduce end-to-end neural network architectures for data assimilation. It comprises two key components: a variational model and a gradient-based solver both implemented as neural networks. A key feature of the proposed end-to-end learning architecture is that we may train the NN models using both supervised and unsupervised strategies. Our numerical experiments on Lorenz-63 and Lorenz-96 systems report significant gain w.r.t. a classic gradient-based minimization of the variational cost both in terms of reconstruction performance and optimization complexity. Intriguingly, we also show that the variational models issued from the true Lorenz-63 and Lorenz-96 ODE representations may not lead to the best reconstruction performance. We believe these results may open new research avenues for the specification of assimilation models in geoscience." @default.
- W3204279182 created "2021-10-11" @default.
- W3204279182 creator A5018721311 @default.
- W3204279182 creator A5030463184 @default.
- W3204279182 creator A5045707906 @default.
- W3204279182 creator A5046811680 @default.
- W3204279182 creator A5073132582 @default.
- W3204279182 creator A5075497720 @default.
- W3204279182 date "2021-10-01" @default.
- W3204279182 modified "2023-10-14" @default.
- W3204279182 title "Learning Variational Data Assimilation Models and Solvers" @default.
- W3204279182 cites W1510319310 @default.
- W3204279182 cites W1915360731 @default.
- W3204279182 cites W1966243290 @default.
- W3204279182 cites W1966764026 @default.
- W3204279182 cites W2124192278 @default.
- W3204279182 cites W2141394518 @default.
- W3204279182 cites W2144144709 @default.
- W3204279182 cites W2145680191 @default.
- W3204279182 cites W2239232218 @default.
- W3204279182 cites W2563710438 @default.
- W3204279182 cites W2748401178 @default.
- W3204279182 cites W2777652944 @default.
- W3204279182 cites W2782977076 @default.
- W3204279182 cites W2899283552 @default.
- W3204279182 cites W2901009362 @default.
- W3204279182 cites W2919115771 @default.
- W3204279182 cites W2952046917 @default.
- W3204279182 cites W2981906852 @default.
- W3204279182 cites W3098900881 @default.
- W3204279182 cites W3101082668 @default.
- W3204279182 cites W3162258555 @default.
- W3204279182 cites W3177453881 @default.
- W3204279182 cites W4231204432 @default.
- W3204279182 cites W4232350701 @default.
- W3204279182 doi "https://doi.org/10.1029/2021ms002572" @default.
- W3204279182 hasPublicationYear "2021" @default.
- W3204279182 type Work @default.
- W3204279182 sameAs 3204279182 @default.
- W3204279182 citedByCount "12" @default.
- W3204279182 countsByYear W32042791822021 @default.
- W3204279182 countsByYear W32042791822022 @default.
- W3204279182 countsByYear W32042791822023 @default.
- W3204279182 crossrefType "journal-article" @default.
- W3204279182 hasAuthorship W3204279182A5018721311 @default.
- W3204279182 hasAuthorship W3204279182A5030463184 @default.
- W3204279182 hasAuthorship W3204279182A5045707906 @default.
- W3204279182 hasAuthorship W3204279182A5046811680 @default.
- W3204279182 hasAuthorship W3204279182A5073132582 @default.
- W3204279182 hasAuthorship W3204279182A5075497720 @default.
- W3204279182 hasBestOaLocation W32042791822 @default.
- W3204279182 hasConcept C108583219 @default.
- W3204279182 hasConcept C11413529 @default.
- W3204279182 hasConcept C119857082 @default.
- W3204279182 hasConcept C121332964 @default.
- W3204279182 hasConcept C126255220 @default.
- W3204279182 hasConcept C147764199 @default.
- W3204279182 hasConcept C153294291 @default.
- W3204279182 hasConcept C154945302 @default.
- W3204279182 hasConcept C199360897 @default.
- W3204279182 hasConcept C24552861 @default.
- W3204279182 hasConcept C2778770139 @default.
- W3204279182 hasConcept C28826006 @default.
- W3204279182 hasConcept C33923547 @default.
- W3204279182 hasConcept C34862557 @default.
- W3204279182 hasConcept C41008148 @default.
- W3204279182 hasConcept C50644808 @default.
- W3204279182 hasConceptScore W3204279182C108583219 @default.
- W3204279182 hasConceptScore W3204279182C11413529 @default.
- W3204279182 hasConceptScore W3204279182C119857082 @default.
- W3204279182 hasConceptScore W3204279182C121332964 @default.
- W3204279182 hasConceptScore W3204279182C126255220 @default.
- W3204279182 hasConceptScore W3204279182C147764199 @default.
- W3204279182 hasConceptScore W3204279182C153294291 @default.
- W3204279182 hasConceptScore W3204279182C154945302 @default.
- W3204279182 hasConceptScore W3204279182C199360897 @default.
- W3204279182 hasConceptScore W3204279182C24552861 @default.
- W3204279182 hasConceptScore W3204279182C2778770139 @default.
- W3204279182 hasConceptScore W3204279182C28826006 @default.
- W3204279182 hasConceptScore W3204279182C33923547 @default.
- W3204279182 hasConceptScore W3204279182C34862557 @default.
- W3204279182 hasConceptScore W3204279182C41008148 @default.
- W3204279182 hasConceptScore W3204279182C50644808 @default.
- W3204279182 hasIssue "10" @default.
- W3204279182 hasLocation W32042791821 @default.
- W3204279182 hasLocation W32042791822 @default.
- W3204279182 hasLocation W32042791823 @default.
- W3204279182 hasLocation W32042791824 @default.
- W3204279182 hasLocation W32042791825 @default.
- W3204279182 hasLocation W32042791826 @default.
- W3204279182 hasLocation W32042791827 @default.
- W3204279182 hasLocation W32042791828 @default.
- W3204279182 hasLocation W32042791829 @default.
- W3204279182 hasOpenAccess W3204279182 @default.
- W3204279182 hasPrimaryLocation W32042791821 @default.
- W3204279182 hasRelatedWork W2048742619 @default.
- W3204279182 hasRelatedWork W2362647823 @default.
- W3204279182 hasRelatedWork W2363680170 @default.
- W3204279182 hasRelatedWork W2371448224 @default.
- W3204279182 hasRelatedWork W2376428685 @default.