FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2900533563> ?p ?o ?g. }

• W2900533563 abstract "Much effort has been devoted to evaluate whether multi-task learning can be leveraged to learn rich representations that can be used in various Natural Language Processing (NLP) down-stream applications. However, there is still a lack of understanding of the settings in which multi-task learning has a significant effect. In this work, we introduce a hierarchical model trained in a multi-task learning setup on a set of carefully selected semantic tasks. The model is trained in a hierarchical fashion to introduce an inductive bias by supervising a set of low level tasks at the bottom layers of the model and more complex tasks at the top layers of the model. This model achieves state-of-the-art results on a number of tasks, namely Named Entity Recognition, Entity Mention Detection and Relation Extraction without hand-engineered features or external NLP tools like syntactic parsers. The hierarchical training supervision induces a set of shared semantic representations at lower layers of the model. We show that as we move from the bottom to the top layers of the model, the hidden states of the layers tend to represent more complex semantic information." @default.
• W2900533563 created "2018-11-29" @default.
• W2900533563 creator A5002672417 @default.
• W2900533563 creator A5037310413 @default.
• W2900533563 creator A5049805631 @default.
• W2900533563 date "2018-11-14" @default.
• W2900533563 modified "2023-09-22" @default.
• W2900533563 title "A Hierarchical Multi-task Approach for Learning Embeddings from Semantic Tasks" @default.
• W2900533563 cites W2060277733 @default.
• W2900533563 cites W2103076621 @default.
• W2900533563 cites W2124700572 @default.
• W2900533563 cites W2134033474 @default.
• W2900533563 cites W2147880316 @default.
• W2900533563 cites W2155069789 @default.
• W2900533563 cites W2155247542 @default.
• W2900533563 cites W2179519966 @default.
• W2900533563 cites W2229639163 @default.
• W2900533563 cites W2250539671 @default.
• W2900533563 cites W2251035762 @default.
• W2900533563 cites W2252031764 @default.
• W2900533563 cites W2293004735 @default.
• W2900533563 cites W2296073425 @default.
• W2900533563 cites W2296283641 @default.
• W2900533563 cites W2407338347 @default.
• W2900533563 cites W2516255829 @default.
• W2900533563 cites W2556468274 @default.
• W2900533563 cites W2610858497 @default.
• W2900533563 cites W2624871570 @default.
• W2900533563 cites W2738152205 @default.
• W2900533563 cites W2741956709 @default.
• W2900533563 cites W2752172973 @default.
• W2900533563 cites W2784231336 @default.
• W2900533563 cites W2786685006 @default.
• W2900533563 cites W2787560479 @default.
• W2900533563 cites W2797498228 @default.
• W2900533563 cites W2799124508 @default.
• W2900533563 cites W2799125718 @default.
• W2900533563 cites W2809324505 @default.
• W2900533563 cites W2913340405 @default.
• W2900533563 cites W2949563612 @default.
• W2900533563 cites W2950726992 @default.
• W2900533563 cites W2963140597 @default.
• W2900533563 cites W2963918774 @default.
• W2900533563 cites W2963925965 @default.
• W2900533563 cites W2964094426 @default.
• W2900533563 cites W3209042722 @default.
• W2900533563 cites W36434594 @default.
• W2900533563 hasPublicationYear "2018" @default.
• W2900533563 type Work @default.
• W2900533563 sameAs 2900533563 @default.
• W2900533563 citedByCount "14" @default.
• W2900533563 countsByYear W29005335632019 @default.
• W2900533563 countsByYear W29005335632020 @default.
• W2900533563 countsByYear W29005335632021 @default.
• W2900533563 crossrefType "posted-content" @default.
• W2900533563 hasAuthorship W2900533563A5002672417 @default.
• W2900533563 hasAuthorship W2900533563A5037310413 @default.
• W2900533563 hasAuthorship W2900533563A5049805631 @default.
• W2900533563 hasConcept C115903868 @default.
• W2900533563 hasConcept C119857082 @default.
• W2900533563 hasConcept C124101348 @default.
• W2900533563 hasConcept C135798126 @default.
• W2900533563 hasConcept C144986985 @default.
• W2900533563 hasConcept C153604712 @default.
• W2900533563 hasConcept C154945302 @default.
• W2900533563 hasConcept C162324750 @default.
• W2900533563 hasConcept C177264268 @default.
• W2900533563 hasConcept C186644900 @default.
• W2900533563 hasConcept C187736073 @default.
• W2900533563 hasConcept C195324797 @default.
• W2900533563 hasConcept C195807954 @default.
• W2900533563 hasConcept C199360897 @default.
• W2900533563 hasConcept C204321447 @default.
• W2900533563 hasConcept C2779439875 @default.
• W2900533563 hasConcept C2780451532 @default.
• W2900533563 hasConcept C41008148 @default.
• W2900533563 hasConceptScore W2900533563C115903868 @default.
• W2900533563 hasConceptScore W2900533563C119857082 @default.
• W2900533563 hasConceptScore W2900533563C124101348 @default.
• W2900533563 hasConceptScore W2900533563C135798126 @default.
• W2900533563 hasConceptScore W2900533563C144986985 @default.
• W2900533563 hasConceptScore W2900533563C153604712 @default.
• W2900533563 hasConceptScore W2900533563C154945302 @default.
• W2900533563 hasConceptScore W2900533563C162324750 @default.
• W2900533563 hasConceptScore W2900533563C177264268 @default.
• W2900533563 hasConceptScore W2900533563C186644900 @default.
• W2900533563 hasConceptScore W2900533563C187736073 @default.
• W2900533563 hasConceptScore W2900533563C195324797 @default.
• W2900533563 hasConceptScore W2900533563C195807954 @default.
• W2900533563 hasConceptScore W2900533563C199360897 @default.
• W2900533563 hasConceptScore W2900533563C204321447 @default.
• W2900533563 hasConceptScore W2900533563C2779439875 @default.
• W2900533563 hasConceptScore W2900533563C2780451532 @default.
• W2900533563 hasConceptScore W2900533563C41008148 @default.
• W2900533563 hasOpenAccess W2900533563 @default.
• W2900533563 hasRelatedWork W168804263 @default.
• W2900533563 hasRelatedWork W1975759955 @default.
• W2900533563 hasRelatedWork W2006710607 @default.
• W2900533563 hasRelatedWork W2250539671 @default.
• W2900533563 hasRelatedWork W2251013762 @default.

• next