Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949680570> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2949680570 abstract "Several mechanisms to focus attention of a neural network on selected parts of its input or memory have been used successfully in deep learning models in recent years. Attention has improved image classification, image captioning, speech recognition, generative models, and learning algorithmic tasks, but it had probably the largest impact on neural machine translation. Recently, similar improvements have been obtained using alternative mechanisms that do not focus on a single part of a memory but operate on all of it in parallel, in a uniform way. Such mechanism, which we call active memory, improved over attention in algorithmic tasks, image processing, and in generative modelling. So far, however, active memory has not improved over attention for most natural language processing tasks, in particular for machine translation. We analyze this shortcoming in this paper and propose an extended model of active memory that matches existing attention models on neural machine translation and generalizes better to longer sentences. We investigate this model and explain why previous active memory models did not succeed. Finally, we discuss when active memory brings most benefits and where attention can be a better choice." @default.
- W2949680570 created "2019-06-27" @default.
- W2949680570 creator A5017529415 @default.
- W2949680570 creator A5031789995 @default.
- W2949680570 date "2016-10-27" @default.
- W2949680570 modified "2023-09-27" @default.
- W2949680570 title "Can Active Memory Replace Attention" @default.
- W2949680570 cites W1485009520 @default.
- W2949680570 cites W1514535095 @default.
- W2949680570 cites W1522301498 @default.
- W2949680570 cites W1753482797 @default.
- W2949680570 cites W2119717200 @default.
- W2949680570 cites W2132043663 @default.
- W2949680570 cites W2147768505 @default.
- W2949680570 cites W2194775991 @default.
- W2949680570 cites W2271840356 @default.
- W2949680570 cites W2337199865 @default.
- W2949680570 cites W2410539690 @default.
- W2949680570 cites W2949245006 @default.
- W2949680570 cites W2949626814 @default.
- W2949680570 cites W2962965465 @default.
- W2949680570 cites W2963069010 @default.
- W2949680570 cites W2963187627 @default.
- W2949680570 cites W2964199361 @default.
- W2949680570 cites W2964308564 @default.
- W2949680570 hasPublicationYear "2016" @default.
- W2949680570 type Work @default.
- W2949680570 sameAs 2949680570 @default.
- W2949680570 citedByCount "1" @default.
- W2949680570 countsByYear W29496805702019 @default.
- W2949680570 crossrefType "posted-content" @default.
- W2949680570 hasAuthorship W2949680570A5017529415 @default.
- W2949680570 hasAuthorship W2949680570A5031789995 @default.
- W2949680570 hasConcept C104317684 @default.
- W2949680570 hasConcept C105580179 @default.
- W2949680570 hasConcept C115961682 @default.
- W2949680570 hasConcept C119857082 @default.
- W2949680570 hasConcept C120665830 @default.
- W2949680570 hasConcept C121332964 @default.
- W2949680570 hasConcept C149364088 @default.
- W2949680570 hasConcept C154945302 @default.
- W2949680570 hasConcept C157657479 @default.
- W2949680570 hasConcept C167966045 @default.
- W2949680570 hasConcept C185592680 @default.
- W2949680570 hasConcept C192209626 @default.
- W2949680570 hasConcept C203005215 @default.
- W2949680570 hasConcept C204321447 @default.
- W2949680570 hasConcept C39890363 @default.
- W2949680570 hasConcept C41008148 @default.
- W2949680570 hasConcept C50644808 @default.
- W2949680570 hasConcept C55493867 @default.
- W2949680570 hasConceptScore W2949680570C104317684 @default.
- W2949680570 hasConceptScore W2949680570C105580179 @default.
- W2949680570 hasConceptScore W2949680570C115961682 @default.
- W2949680570 hasConceptScore W2949680570C119857082 @default.
- W2949680570 hasConceptScore W2949680570C120665830 @default.
- W2949680570 hasConceptScore W2949680570C121332964 @default.
- W2949680570 hasConceptScore W2949680570C149364088 @default.
- W2949680570 hasConceptScore W2949680570C154945302 @default.
- W2949680570 hasConceptScore W2949680570C157657479 @default.
- W2949680570 hasConceptScore W2949680570C167966045 @default.
- W2949680570 hasConceptScore W2949680570C185592680 @default.
- W2949680570 hasConceptScore W2949680570C192209626 @default.
- W2949680570 hasConceptScore W2949680570C203005215 @default.
- W2949680570 hasConceptScore W2949680570C204321447 @default.
- W2949680570 hasConceptScore W2949680570C39890363 @default.
- W2949680570 hasConceptScore W2949680570C41008148 @default.
- W2949680570 hasConceptScore W2949680570C50644808 @default.
- W2949680570 hasConceptScore W2949680570C55493867 @default.
- W2949680570 hasLocation W29496805701 @default.
- W2949680570 hasOpenAccess W2949680570 @default.
- W2949680570 hasPrimaryLocation W29496805701 @default.
- W2949680570 hasRelatedWork W2037573256 @default.
- W2949680570 hasRelatedWork W2400680200 @default.
- W2949680570 hasRelatedWork W2526108667 @default.
- W2949680570 hasRelatedWork W2545625743 @default.
- W2949680570 hasRelatedWork W2558263348 @default.
- W2949680570 hasRelatedWork W2591866828 @default.
- W2949680570 hasRelatedWork W2755728112 @default.
- W2949680570 hasRelatedWork W2795429538 @default.
- W2949680570 hasRelatedWork W2944805769 @default.
- W2949680570 hasRelatedWork W2952136670 @default.
- W2949680570 hasRelatedWork W2952191002 @default.
- W2949680570 hasRelatedWork W2953542613 @default.
- W2949680570 hasRelatedWork W3004860304 @default.
- W2949680570 hasRelatedWork W3036615999 @default.
- W2949680570 hasRelatedWork W3109885824 @default.
- W2949680570 hasRelatedWork W3132023005 @default.
- W2949680570 hasRelatedWork W3138923709 @default.
- W2949680570 hasRelatedWork W3178829432 @default.
- W2949680570 hasRelatedWork W3211695811 @default.
- W2949680570 hasRelatedWork W3015492205 @default.
- W2949680570 isParatext "false" @default.
- W2949680570 isRetracted "false" @default.
- W2949680570 magId "2949680570" @default.
- W2949680570 workType "article" @default.