Matches in SemOpenAlex for { <https://semopenalex.org/work/W2981531693> ?p ?o ?g. }
- W2981531693 abstract "Residual networks (ResNets) are a deep learning architecture with the recursive structure [x_{k+1} = x_k + R_k(x_k)] where $R_k$ is a neural network and the copying of the input $x_k$ is called a skip connection. This structure can be seen as the explicit Euler discretisation of an associated ordinary differential equation. We use this interpretation to show that by simultaneously increasing the number of skip connection as well as the expressivity of the networks $R_k$ the flow of an arbitrary right hand side [fin L^1left( I; mathcal C_b^{0, 1}(mathbb R^d; mathbb R^d)right)] can be approximated uniformly by deep ReLU ResNets on compact sets. Further, we derive estimates on the number of parameters needed to do this up to a prescribed accuracy under temporal regularity assumptions. Finally, we discuss the possibility of using ResNets for diffeomorphic matching problems and propose some next steps in the theoretical foundation of this approach." @default.
- W2981531693 created "2019-11-01" @default.
- W2981531693 creator A5012807288 @default.
- W2981531693 date "2019-10-21" @default.
- W2981531693 modified "2023-09-27" @default.
- W2981531693 title "Universal flow approximation with deep residual networks." @default.
- W2981531693 cites W1970346034 @default.
- W2981531693 cites W2007008305 @default.
- W2981531693 cites W2024226655 @default.
- W2981531693 cites W2165612210 @default.
- W2981531693 cites W2410678056 @default.
- W2981531693 cites W2505225485 @default.
- W2981531693 cites W2555745781 @default.
- W2981531693 cites W2588006840 @default.
- W2981531693 cites W2600297185 @default.
- W2981531693 cites W2767286248 @default.
- W2981531693 cites W2787302432 @default.
- W2981531693 cites W2788421913 @default.
- W2981531693 cites W2803629276 @default.
- W2981531693 cites W2808746463 @default.
- W2981531693 cites W2810898455 @default.
- W2981531693 cites W2890291741 @default.
- W2981531693 cites W2890889625 @default.
- W2981531693 cites W2898318847 @default.
- W2981531693 cites W2899507754 @default.
- W2981531693 cites W2921044119 @default.
- W2981531693 cites W2926003515 @default.
- W2981531693 cites W2933916181 @default.
- W2981531693 cites W2944900050 @default.
- W2981531693 cites W2947179935 @default.
- W2981531693 cites W2951548721 @default.
- W2981531693 cites W2954833584 @default.
- W2981531693 cites W2955831823 @default.
- W2981531693 cites W2963146412 @default.
- W2981531693 cites W2963302407 @default.
- W2981531693 cites W2969755609 @default.
- W2981531693 cites W2980448402 @default.
- W2981531693 cites W3008349919 @default.
- W2981531693 cites W3025750359 @default.
- W2981531693 cites W3098011980 @default.
- W2981531693 cites W3104183394 @default.
- W2981531693 hasPublicationYear "2019" @default.
- W2981531693 type Work @default.
- W2981531693 sameAs 2981531693 @default.
- W2981531693 citedByCount "1" @default.
- W2981531693 countsByYear W29815316932021 @default.
- W2981531693 crossrefType "posted-content" @default.
- W2981531693 hasAuthorship W2981531693A5012807288 @default.
- W2981531693 hasConcept C105795698 @default.
- W2981531693 hasConcept C11413529 @default.
- W2981531693 hasConcept C118615104 @default.
- W2981531693 hasConcept C13355873 @default.
- W2981531693 hasConcept C134306372 @default.
- W2981531693 hasConcept C155512373 @default.
- W2981531693 hasConcept C165064840 @default.
- W2981531693 hasConcept C199360897 @default.
- W2981531693 hasConcept C202444582 @default.
- W2981531693 hasConcept C2524010 @default.
- W2981531693 hasConcept C28826006 @default.
- W2981531693 hasConcept C33923547 @default.
- W2981531693 hasConcept C38349280 @default.
- W2981531693 hasConcept C41008148 @default.
- W2981531693 hasConcept C51544822 @default.
- W2981531693 hasConcept C527412718 @default.
- W2981531693 hasConcept C62884695 @default.
- W2981531693 hasConcept C73000952 @default.
- W2981531693 hasConcept C78045399 @default.
- W2981531693 hasConceptScore W2981531693C105795698 @default.
- W2981531693 hasConceptScore W2981531693C11413529 @default.
- W2981531693 hasConceptScore W2981531693C118615104 @default.
- W2981531693 hasConceptScore W2981531693C13355873 @default.
- W2981531693 hasConceptScore W2981531693C134306372 @default.
- W2981531693 hasConceptScore W2981531693C155512373 @default.
- W2981531693 hasConceptScore W2981531693C165064840 @default.
- W2981531693 hasConceptScore W2981531693C199360897 @default.
- W2981531693 hasConceptScore W2981531693C202444582 @default.
- W2981531693 hasConceptScore W2981531693C2524010 @default.
- W2981531693 hasConceptScore W2981531693C28826006 @default.
- W2981531693 hasConceptScore W2981531693C33923547 @default.
- W2981531693 hasConceptScore W2981531693C38349280 @default.
- W2981531693 hasConceptScore W2981531693C41008148 @default.
- W2981531693 hasConceptScore W2981531693C51544822 @default.
- W2981531693 hasConceptScore W2981531693C527412718 @default.
- W2981531693 hasConceptScore W2981531693C62884695 @default.
- W2981531693 hasConceptScore W2981531693C73000952 @default.
- W2981531693 hasConceptScore W2981531693C78045399 @default.
- W2981531693 hasLocation W29815316931 @default.
- W2981531693 hasOpenAccess W2981531693 @default.
- W2981531693 hasPrimaryLocation W29815316931 @default.
- W2981531693 hasRelatedWork W1514899150 @default.
- W2981531693 hasRelatedWork W1804987519 @default.
- W2981531693 hasRelatedWork W1969815718 @default.
- W2981531693 hasRelatedWork W1976537103 @default.
- W2981531693 hasRelatedWork W1992650621 @default.
- W2981531693 hasRelatedWork W2012411663 @default.
- W2981531693 hasRelatedWork W2132492300 @default.
- W2981531693 hasRelatedWork W2135596970 @default.
- W2981531693 hasRelatedWork W2144381961 @default.
- W2981531693 hasRelatedWork W2468975740 @default.
- W2981531693 hasRelatedWork W2787302432 @default.