FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2969856879> ?p ?o ?g. }

• W2969856879 endingPage "154166" @default.
• W2969856879 startingPage "154157" @default.
• W2969856879 abstract "Neural Machine Translation(NMT) has achieved notable results in high-resource languages, but still works poorly on low-resource languages. As times goes on, It is widely recognized that transfer learning methods are effective for low-resource language problems. However, existing transfer learning methods are typically based on the parent-child architecture, which does not adequately take advantages of helpful languages. In this paper, inspired by human transitive inference and learning ability, we handle this issue by proposing a new hierarchical transfer learning architecture for low-resource languages. In the architecture, the NMT model is trained in the unrelated high-resource language pair, the similar intermediate language pair and the low-resource language pair in turn. Correspondingly, the parameters are transferred and fine-tuned layer by layer for initialization. In this way, our hierarchical transfer learning architecture simultaneously combines the data volume advantages of high-resource languages and the syntactic similarity advantages of cognate languages. Specially, we utilize Byte Pair Encoding(BPE) and character-level embedding for data pre-processing, which effectively solve the problem of out of vocabulary(OOV). Experimental results on Uygur-Chinese and Turkish-English translation demonstrate the superiorities of the proposed architecture over the NMT model with parent-child architecture." @default.
• W2969856879 created "2019-08-29" @default.
• W2969856879 creator A5012997528 @default.
• W2969856879 creator A5032966156 @default.
• W2969856879 creator A5034313244 @default.
• W2969856879 creator A5050270773 @default.
• W2969856879 creator A5068793737 @default.
• W2969856879 date "2019-01-01" @default.
• W2969856879 modified "2023-10-12" @default.
• W2969856879 title "Hierarchical Transfer Learning Architecture for Low-Resource Neural Machine Translation" @default.
• W2969856879 cites W1902237438 @default.
• W2969856879 cites W1985697265 @default.
• W2969856879 cites W1988835709 @default.
• W2969856879 cites W2124807415 @default.
• W2969856879 cites W2133601766 @default.
• W2969856879 cites W2165698076 @default.
• W2969856879 cites W2214409633 @default.
• W2969856879 cites W2251743902 @default.
• W2969856879 cites W2341914330 @default.
• W2969856879 cites W2561274697 @default.
• W2969856879 cites W2760656271 @default.
• W2969856879 cites W2761615849 @default.
• W2969856879 cites W2762776925 @default.
• W2969856879 cites W2796108585 @default.
• W2969856879 cites W2888541716 @default.
• W2969856879 cites W2962696859 @default.
• W2969856879 cites W2962784628 @default.
• W2969856879 cites W2963088995 @default.
• W2969856879 cites W2963168418 @default.
• W2969856879 cites W2963216553 @default.
• W2969856879 cites W2963247703 @default.
• W2969856879 cites W2963251942 @default.
• W2969856879 cites W2963331137 @default.
• W2969856879 cites W2963357083 @default.
• W2969856879 cites W2963506925 @default.
• W2969856879 cites W2963939565 @default.
• W2969856879 cites W3098341425 @default.
• W2969856879 cites W3104652516 @default.
• W2969856879 cites W4255421341 @default.
• W2969856879 doi "https://doi.org/10.1109/access.2019.2936002" @default.
• W2969856879 hasPublicationYear "2019" @default.
• W2969856879 type Work @default.
• W2969856879 sameAs 2969856879 @default.
• W2969856879 citedByCount "10" @default.
• W2969856879 countsByYear W29698568792020 @default.
• W2969856879 countsByYear W29698568792021 @default.
• W2969856879 countsByYear W29698568792022 @default.
• W2969856879 countsByYear W29698568792023 @default.
• W2969856879 crossrefType "journal-article" @default.
• W2969856879 hasAuthorship W2969856879A5012997528 @default.
• W2969856879 hasAuthorship W2969856879A5032966156 @default.
• W2969856879 hasAuthorship W2969856879A5034313244 @default.
• W2969856879 hasAuthorship W2969856879A5050270773 @default.
• W2969856879 hasAuthorship W2969856879A5068793737 @default.
• W2969856879 hasBestOaLocation W29698568791 @default.
• W2969856879 hasConcept C104317684 @default.
• W2969856879 hasConcept C105580179 @default.
• W2969856879 hasConcept C123657996 @default.
• W2969856879 hasConcept C142362112 @default.
• W2969856879 hasConcept C149364088 @default.
• W2969856879 hasConcept C150899416 @default.
• W2969856879 hasConcept C153349607 @default.
• W2969856879 hasConcept C154945302 @default.
• W2969856879 hasConcept C185592680 @default.
• W2969856879 hasConcept C203005215 @default.
• W2969856879 hasConcept C204321447 @default.
• W2969856879 hasConcept C206345919 @default.
• W2969856879 hasConcept C31258907 @default.
• W2969856879 hasConcept C41008148 @default.
• W2969856879 hasConcept C55493867 @default.
• W2969856879 hasConceptScore W2969856879C104317684 @default.
• W2969856879 hasConceptScore W2969856879C105580179 @default.
• W2969856879 hasConceptScore W2969856879C123657996 @default.
• W2969856879 hasConceptScore W2969856879C142362112 @default.
• W2969856879 hasConceptScore W2969856879C149364088 @default.
• W2969856879 hasConceptScore W2969856879C150899416 @default.
• W2969856879 hasConceptScore W2969856879C153349607 @default.
• W2969856879 hasConceptScore W2969856879C154945302 @default.
• W2969856879 hasConceptScore W2969856879C185592680 @default.
• W2969856879 hasConceptScore W2969856879C203005215 @default.
• W2969856879 hasConceptScore W2969856879C204321447 @default.
• W2969856879 hasConceptScore W2969856879C206345919 @default.
• W2969856879 hasConceptScore W2969856879C31258907 @default.
• W2969856879 hasConceptScore W2969856879C41008148 @default.
• W2969856879 hasConceptScore W2969856879C55493867 @default.
• W2969856879 hasLocation W29698568791 @default.
• W2969856879 hasLocation W29698568792 @default.
• W2969856879 hasOpenAccess W2969856879 @default.
• W2969856879 hasPrimaryLocation W29698568791 @default.
• W2969856879 hasRelatedWork W2728761353 @default.
• W2969856879 hasRelatedWork W2807006873 @default.
• W2969856879 hasRelatedWork W2883671469 @default.
• W2969856879 hasRelatedWork W2962780935 @default.
• W2969856879 hasRelatedWork W2990482163 @default.
• W2969856879 hasRelatedWork W3011059803 @default.
• W2969856879 hasRelatedWork W3151736118 @default.
• W2969856879 hasRelatedWork W3201126466 @default.
• W2969856879 hasRelatedWork W4287236333 @default.

• next