Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366985237> ?p ?o ?g. }
Showing items 1 to 53 of
53
with 100 items per page.
- W4366985237 abstract "<p>This paper presents a concise mathematical framework for investigating both feed-forward and backward process, during the training to learn model weights, of an artificial neural network (ANN). Inspired from the idea of the two-step rule for backpropagation, we define a notion of F-adjoint which is aimed at a better description of the backpropagation algorithm. In particular, by introducing the notions of F-propagation and F-adjoint through a deep neural network architecture, the backpropagation associated to a cost/loss function is proven to be completely characterized by the F-adjoint of the corresponding F-propagation relatively to the partial derivative, with respect to the inputs, of the cost function.</p>" @default.
- W4366985237 created "2023-04-27" @default.
- W4366985237 creator A5085909004 @default.
- W4366985237 date "2023-04-25" @default.
- W4366985237 modified "2023-09-29" @default.
- W4366985237 title "Backpropagation and F-adjoint" @default.
- W4366985237 doi "https://doi.org/10.36227/techrxiv.22650193.v1" @default.
- W4366985237 hasPublicationYear "2023" @default.
- W4366985237 type Work @default.
- W4366985237 citedByCount "0" @default.
- W4366985237 crossrefType "posted-content" @default.
- W4366985237 hasAuthorship W4366985237A5085909004 @default.
- W4366985237 hasBestOaLocation W43669852371 @default.
- W4366985237 hasConcept C111919701 @default.
- W4366985237 hasConcept C11413529 @default.
- W4366985237 hasConcept C14036430 @default.
- W4366985237 hasConcept C154945302 @default.
- W4366985237 hasConcept C155032097 @default.
- W4366985237 hasConcept C28826006 @default.
- W4366985237 hasConcept C33923547 @default.
- W4366985237 hasConcept C41008148 @default.
- W4366985237 hasConcept C50644808 @default.
- W4366985237 hasConcept C78458016 @default.
- W4366985237 hasConcept C86803240 @default.
- W4366985237 hasConcept C98045186 @default.
- W4366985237 hasConceptScore W4366985237C111919701 @default.
- W4366985237 hasConceptScore W4366985237C11413529 @default.
- W4366985237 hasConceptScore W4366985237C14036430 @default.
- W4366985237 hasConceptScore W4366985237C154945302 @default.
- W4366985237 hasConceptScore W4366985237C155032097 @default.
- W4366985237 hasConceptScore W4366985237C28826006 @default.
- W4366985237 hasConceptScore W4366985237C33923547 @default.
- W4366985237 hasConceptScore W4366985237C41008148 @default.
- W4366985237 hasConceptScore W4366985237C50644808 @default.
- W4366985237 hasConceptScore W4366985237C78458016 @default.
- W4366985237 hasConceptScore W4366985237C86803240 @default.
- W4366985237 hasConceptScore W4366985237C98045186 @default.
- W4366985237 hasLocation W43669852371 @default.
- W4366985237 hasOpenAccess W4366985237 @default.
- W4366985237 hasPrimaryLocation W43669852371 @default.
- W4366985237 hasRelatedWork W2040019480 @default.
- W4366985237 hasRelatedWork W2105899231 @default.
- W4366985237 hasRelatedWork W2350832155 @default.
- W4366985237 hasRelatedWork W2362189222 @default.
- W4366985237 hasRelatedWork W2383491577 @default.
- W4366985237 hasRelatedWork W2391384657 @default.
- W4366985237 hasRelatedWork W2435007635 @default.
- W4366985237 hasRelatedWork W4304136584 @default.
- W4366985237 hasRelatedWork W985376067 @default.
- W4366985237 hasRelatedWork W2516580779 @default.
- W4366985237 isParatext "false" @default.
- W4366985237 isRetracted "false" @default.
- W4366985237 workType "article" @default.