FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3189781688> ?p ?o ?g. }

• W3189781688 endingPage "184" @default.
• W3189781688 startingPage "164" @default.
• W3189781688 abstract "Abstract We introduce a novel algorithm for estimating optimal parameters of linearized assignment flows for image labeling. An exact formula is derived for the parameter gradient of any loss function that is constrained by the linear system of ODEs determining the linearized assignment flow. We show how to efficiently evaluate this formula using a Krylov subspace and a low-rank approximation. This enables us to perform parameter learning by Riemannian gradient descent in the parameter space, without the need to backpropagate errors or to solve an adjoint equation. Experiments demonstrate that our method performs as good as highly-tuned machine learning software using automatic differentiation. Unlike methods employing automatic differentiation, our approach yields a low-dimensional representation of internal parameters and their dynamics which helps to understand how assignment flows and more generally neural networks work and perform." @default.
• W3189781688 created "2021-08-16" @default.
• W3189781688 creator A5039507181 @default.
• W3189781688 creator A5082060792 @default.
• W3189781688 creator A5088815590 @default.
• W3189781688 date "2023-01-01" @default.
• W3189781688 modified "2023-10-13" @default.
• W3189781688 title "Learning Linearized Assignment Flows for Image Labeling" @default.
• W3189781688 cites W1483804921 @default.
• W3189781688 cites W1488435683 @default.
• W3189781688 cites W1506342804 @default.
• W3189781688 cites W1521738998 @default.
• W3189781688 cites W1540455107 @default.
• W3189781688 cites W1984347942 @default.
• W3189781688 cites W2018931076 @default.
• W3189781688 cites W2028212620 @default.
• W3189781688 cites W2057602562 @default.
• W3189781688 cites W2061356302 @default.
• W3189781688 cites W2073389244 @default.
• W3189781688 cites W2077852010 @default.
• W3189781688 cites W2106565812 @default.
• W3189781688 cites W2300087321 @default.
• W3189781688 cites W2480854438 @default.
• W3189781688 cites W2990578246 @default.
• W3189781688 cites W2993076031 @default.
• W3189781688 cites W3159709631 @default.
• W3189781688 cites W3167416242 @default.
• W3189781688 cites W3215925321 @default.
• W3189781688 cites W4211077197 @default.
• W3189781688 cites W4247163269 @default.
• W3189781688 doi "https://doi.org/10.1007/s10851-022-01132-9" @default.
• W3189781688 hasPublicationYear "2023" @default.
• W3189781688 type Work @default.
• W3189781688 sameAs 3189781688 @default.
• W3189781688 citedByCount "2" @default.
• W3189781688 countsByYear W31897816882020 @default.
• W3189781688 countsByYear W31897816882023 @default.
• W3189781688 crossrefType "journal-article" @default.
• W3189781688 hasAuthorship W3189781688A5039507181 @default.
• W3189781688 hasAuthorship W3189781688A5082060792 @default.
• W3189781688 hasAuthorship W3189781688A5088815590 @default.
• W3189781688 hasBestOaLocation W31897816881 @default.
• W3189781688 hasConcept C11413529 @default.
• W3189781688 hasConcept C114614502 @default.
• W3189781688 hasConcept C115961682 @default.
• W3189781688 hasConcept C126255220 @default.
• W3189781688 hasConcept C133512626 @default.
• W3189781688 hasConcept C134306372 @default.
• W3189781688 hasConcept C14036430 @default.
• W3189781688 hasConcept C147060835 @default.
• W3189781688 hasConcept C153258448 @default.
• W3189781688 hasConcept C154945302 @default.
• W3189781688 hasConcept C159694833 @default.
• W3189781688 hasConcept C164226766 @default.
• W3189781688 hasConcept C167879884 @default.
• W3189781688 hasConcept C17744445 @default.
• W3189781688 hasConcept C199539241 @default.
• W3189781688 hasConcept C2524010 @default.
• W3189781688 hasConcept C2776359362 @default.
• W3189781688 hasConcept C28826006 @default.
• W3189781688 hasConcept C32834561 @default.
• W3189781688 hasConcept C33923547 @default.
• W3189781688 hasConcept C38349280 @default.
• W3189781688 hasConcept C41008148 @default.
• W3189781688 hasConcept C45374587 @default.
• W3189781688 hasConcept C50644808 @default.
• W3189781688 hasConcept C73586568 @default.
• W3189781688 hasConcept C78458016 @default.
• W3189781688 hasConcept C86803240 @default.
• W3189781688 hasConcept C94625758 @default.
• W3189781688 hasConceptScore W3189781688C11413529 @default.
• W3189781688 hasConceptScore W3189781688C114614502 @default.
• W3189781688 hasConceptScore W3189781688C115961682 @default.
• W3189781688 hasConceptScore W3189781688C126255220 @default.
• W3189781688 hasConceptScore W3189781688C133512626 @default.
• W3189781688 hasConceptScore W3189781688C134306372 @default.
• W3189781688 hasConceptScore W3189781688C14036430 @default.
• W3189781688 hasConceptScore W3189781688C147060835 @default.
• W3189781688 hasConceptScore W3189781688C153258448 @default.
• W3189781688 hasConceptScore W3189781688C154945302 @default.
• W3189781688 hasConceptScore W3189781688C159694833 @default.
• W3189781688 hasConceptScore W3189781688C164226766 @default.
• W3189781688 hasConceptScore W3189781688C167879884 @default.
• W3189781688 hasConceptScore W3189781688C17744445 @default.
• W3189781688 hasConceptScore W3189781688C199539241 @default.
• W3189781688 hasConceptScore W3189781688C2524010 @default.
• W3189781688 hasConceptScore W3189781688C2776359362 @default.
• W3189781688 hasConceptScore W3189781688C28826006 @default.
• W3189781688 hasConceptScore W3189781688C32834561 @default.
• W3189781688 hasConceptScore W3189781688C33923547 @default.
• W3189781688 hasConceptScore W3189781688C38349280 @default.
• W3189781688 hasConceptScore W3189781688C41008148 @default.
• W3189781688 hasConceptScore W3189781688C45374587 @default.
• W3189781688 hasConceptScore W3189781688C50644808 @default.
• W3189781688 hasConceptScore W3189781688C73586568 @default.
• W3189781688 hasConceptScore W3189781688C78458016 @default.
• W3189781688 hasConceptScore W3189781688C86803240 @default.
• W3189781688 hasConceptScore W3189781688C94625758 @default.

• next