FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4295528346> ?p ?o ?g. }

• W4295528346 endingPage "24" @default.
• W4295528346 startingPage "15" @default.
• W4295528346 abstract "Although deep learning models have been successfully adopted in many applications, they are facing challenges to be deployed on energy-limited devices (e.g., some mobile devices, etc.) due to their high computation complexity. In this paper, we focus on reducing the costs of Gated Recurrent Units (GRUs) for time-series prediction tasks and we propose a new pruning method that can recognize and remove the neural connections that have little influence on the network loss, using a controllable threshold on the absolute value of the pre-trained GRU weights. This is different from existing approaches which usually try to find and preserve the connections with large weight values. We further propose a sparse-connection GRU model (SCGRU) that only needs a one-time pruning (with fine-tuning), rather than using multiple prune-retrain cycles. A large number of experimental results demonstrate that the proposed method is able to largely reduce the storage and computation costs while achieving the state-of-arts performance in two datasets. Code is available ( https://github.com/imLingo/SCGRU)." @default.
• W4295528346 created "2022-09-14" @default.
• W4295528346 creator A5032828285 @default.
• W4295528346 creator A5052480553 @default.
• W4295528346 creator A5063368193 @default.
• W4295528346 creator A5071266785 @default.
• W4295528346 creator A5073361034 @default.
• W4295528346 creator A5085478964 @default.
• W4295528346 date "2022-11-01" @default.
• W4295528346 modified "2023-10-17" @default.
• W4295528346 title "One-shot pruning of gated recurrent unit neural network by sensitivity for time-series prediction" @default.
• W4295528346 cites W2064675550 @default.
• W4295528346 cites W2097117768 @default.
• W4295528346 cites W2097533491 @default.
• W4295528346 cites W2112796928 @default.
• W4295528346 cites W2145085734 @default.
• W4295528346 cites W2145563843 @default.
• W4295528346 cites W2153635508 @default.
• W4295528346 cites W2156297475 @default.
• W4295528346 cites W2162635690 @default.
• W4295528346 cites W2172064003 @default.
• W4295528346 cites W2613328025 @default.
• W4295528346 cites W2621645821 @default.
• W4295528346 cites W2896827527 @default.
• W4295528346 cites W2919115771 @default.
• W4295528346 cites W2991232789 @default.
• W4295528346 cites W3014829561 @default.
• W4295528346 cites W3088389255 @default.
• W4295528346 cites W4246587917 @default.
• W4295528346 doi "https://doi.org/10.1016/j.neucom.2022.09.026" @default.
• W4295528346 hasPublicationYear "2022" @default.
• W4295528346 type Work @default.
• W4295528346 citedByCount "2" @default.
• W4295528346 countsByYear W42955283462023 @default.
• W4295528346 crossrefType "journal-article" @default.
• W4295528346 hasAuthorship W4295528346A5032828285 @default.
• W4295528346 hasAuthorship W4295528346A5052480553 @default.
• W4295528346 hasAuthorship W4295528346A5063368193 @default.
• W4295528346 hasAuthorship W4295528346A5071266785 @default.
• W4295528346 hasAuthorship W4295528346A5073361034 @default.
• W4295528346 hasAuthorship W4295528346A5085478964 @default.
• W4295528346 hasBestOaLocation W42955283461 @default.
• W4295528346 hasConcept C108010975 @default.
• W4295528346 hasConcept C108583219 @default.
• W4295528346 hasConcept C11413529 @default.
• W4295528346 hasConcept C119857082 @default.
• W4295528346 hasConcept C120665830 @default.
• W4295528346 hasConcept C121332964 @default.
• W4295528346 hasConcept C127413603 @default.
• W4295528346 hasConcept C143724316 @default.
• W4295528346 hasConcept C151730666 @default.
• W4295528346 hasConcept C153180895 @default.
• W4295528346 hasConcept C154945302 @default.
• W4295528346 hasConcept C177264268 @default.
• W4295528346 hasConcept C192209626 @default.
• W4295528346 hasConcept C199360897 @default.
• W4295528346 hasConcept C21200559 @default.
• W4295528346 hasConcept C24326235 @default.
• W4295528346 hasConcept C2776760102 @default.
• W4295528346 hasConcept C41008148 @default.
• W4295528346 hasConcept C45374587 @default.
• W4295528346 hasConcept C50644808 @default.
• W4295528346 hasConcept C6557445 @default.
• W4295528346 hasConcept C86803240 @default.
• W4295528346 hasConceptScore W4295528346C108010975 @default.
• W4295528346 hasConceptScore W4295528346C108583219 @default.
• W4295528346 hasConceptScore W4295528346C11413529 @default.
• W4295528346 hasConceptScore W4295528346C119857082 @default.
• W4295528346 hasConceptScore W4295528346C120665830 @default.
• W4295528346 hasConceptScore W4295528346C121332964 @default.
• W4295528346 hasConceptScore W4295528346C127413603 @default.
• W4295528346 hasConceptScore W4295528346C143724316 @default.
• W4295528346 hasConceptScore W4295528346C151730666 @default.
• W4295528346 hasConceptScore W4295528346C153180895 @default.
• W4295528346 hasConceptScore W4295528346C154945302 @default.
• W4295528346 hasConceptScore W4295528346C177264268 @default.
• W4295528346 hasConceptScore W4295528346C192209626 @default.
• W4295528346 hasConceptScore W4295528346C199360897 @default.
• W4295528346 hasConceptScore W4295528346C21200559 @default.
• W4295528346 hasConceptScore W4295528346C24326235 @default.
• W4295528346 hasConceptScore W4295528346C2776760102 @default.
• W4295528346 hasConceptScore W4295528346C41008148 @default.
• W4295528346 hasConceptScore W4295528346C45374587 @default.
• W4295528346 hasConceptScore W4295528346C50644808 @default.
• W4295528346 hasConceptScore W4295528346C6557445 @default.
• W4295528346 hasConceptScore W4295528346C86803240 @default.
• W4295528346 hasFunder F4320321001 @default.
• W4295528346 hasFunder F4320322665 @default.
• W4295528346 hasFunder F4320325626 @default.
• W4295528346 hasFunder F4320336746 @default.
• W4295528346 hasLocation W42955283461 @default.
• W4295528346 hasOpenAccess W4295528346 @default.
• W4295528346 hasPrimaryLocation W42955283461 @default.
• W4295528346 hasRelatedWork W2145253956 @default.
• W4295528346 hasRelatedWork W2355171581 @default.
• W4295528346 hasRelatedWork W2373300491 @default.
• W4295528346 hasRelatedWork W2378744544 @default.
• W4295528346 hasRelatedWork W2379704676 @default.

• next