Matches in SemOpenAlex for { <https://semopenalex.org/work/W3113679918> ?p ?o ?g. }
- W3113679918 abstract "Various methods have already been proposed for learning entity embeddings from text descriptions. Such embeddings are commonly used for inferring properties of entities, for recommendation and entity-oriented search, and for injecting background knowledge into neural architectures, among others. Entity embeddings essentially serve as a compact encoding of a similarity relation, but similarity is an inherently multi-faceted notion. By representing entities as single vectors, existing methods leave it to downstream applications to identify these different facets, and to select the most relevant ones. In this paper, we propose a model that instead learns several vectors for each entity, each of which intuitively captures a different aspect of the considered domain. We use a mixture-of-experts formulation to jointly learn these facet-specific embeddings. The individual entity embeddings are learned using a variant of the GloVe model, which has the advantage that we can easily identify which properties are modelled well in which of the learned embeddings. This is exploited by an associated gating network, which uses pre-trained word vectors to encourage the properties that are modelled by a given embedding to be semantically coherent, i.e. to encourage each of the individual embeddings to capture a meaningful facet." @default.
- W3113679918 created "2021-01-05" @default.
- W3113679918 creator A5009887598 @default.
- W3113679918 creator A5027982939 @default.
- W3113679918 creator A5055360338 @default.
- W3113679918 creator A5059764145 @default.
- W3113679918 date "2020-01-01" @default.
- W3113679918 modified "2023-10-16" @default.
- W3113679918 title "A Mixture-of-Experts Model for Learning Multi-Facet Entity Embeddings" @default.
- W3113679918 cites W1796264214 @default.
- W3113679918 cites W2015547952 @default.
- W3113679918 cites W2082840017 @default.
- W3113679918 cites W2116834771 @default.
- W3113679918 cites W2143017621 @default.
- W3113679918 cites W2147152072 @default.
- W3113679918 cites W2150884987 @default.
- W3113679918 cites W2250539671 @default.
- W3113679918 cites W2277612126 @default.
- W3113679918 cites W2509893387 @default.
- W3113679918 cites W2517063656 @default.
- W3113679918 cites W2738334887 @default.
- W3113679918 cites W2741252866 @default.
- W3113679918 cites W2753738274 @default.
- W3113679918 cites W2899286282 @default.
- W3113679918 cites W2902602021 @default.
- W3113679918 cites W2903538854 @default.
- W3113679918 cites W2913433659 @default.
- W3113679918 cites W2949687090 @default.
- W3113679918 cites W2951048068 @default.
- W3113679918 cites W2962917899 @default.
- W3113679918 cites W2963104724 @default.
- W3113679918 cites W2963226019 @default.
- W3113679918 cites W2963341956 @default.
- W3113679918 cites W2963773425 @default.
- W3113679918 cites W2964127395 @default.
- W3113679918 cites W2964238855 @default.
- W3113679918 cites W2983945350 @default.
- W3113679918 cites W2983995706 @default.
- W3113679918 cites W901252737 @default.
- W3113679918 doi "https://doi.org/10.18653/v1/2020.coling-main.449" @default.
- W3113679918 hasPublicationYear "2020" @default.
- W3113679918 type Work @default.
- W3113679918 sameAs 3113679918 @default.
- W3113679918 citedByCount "3" @default.
- W3113679918 countsByYear W31136799182021 @default.
- W3113679918 crossrefType "proceedings-article" @default.
- W3113679918 hasAuthorship W3113679918A5009887598 @default.
- W3113679918 hasAuthorship W3113679918A5027982939 @default.
- W3113679918 hasAuthorship W3113679918A5055360338 @default.
- W3113679918 hasAuthorship W3113679918A5059764145 @default.
- W3113679918 hasBestOaLocation W31136799181 @default.
- W3113679918 hasConcept C103278499 @default.
- W3113679918 hasConcept C115961682 @default.
- W3113679918 hasConcept C124101348 @default.
- W3113679918 hasConcept C125411270 @default.
- W3113679918 hasConcept C134306372 @default.
- W3113679918 hasConcept C154945302 @default.
- W3113679918 hasConcept C15744967 @default.
- W3113679918 hasConcept C187288502 @default.
- W3113679918 hasConcept C204321447 @default.
- W3113679918 hasConcept C2524010 @default.
- W3113679918 hasConcept C25343380 @default.
- W3113679918 hasConcept C2865642 @default.
- W3113679918 hasConcept C33923547 @default.
- W3113679918 hasConcept C36503486 @default.
- W3113679918 hasConcept C41008148 @default.
- W3113679918 hasConcept C41608201 @default.
- W3113679918 hasConcept C43122875 @default.
- W3113679918 hasConcept C77805123 @default.
- W3113679918 hasConcept C80444323 @default.
- W3113679918 hasConcept C90805587 @default.
- W3113679918 hasConceptScore W3113679918C103278499 @default.
- W3113679918 hasConceptScore W3113679918C115961682 @default.
- W3113679918 hasConceptScore W3113679918C124101348 @default.
- W3113679918 hasConceptScore W3113679918C125411270 @default.
- W3113679918 hasConceptScore W3113679918C134306372 @default.
- W3113679918 hasConceptScore W3113679918C154945302 @default.
- W3113679918 hasConceptScore W3113679918C15744967 @default.
- W3113679918 hasConceptScore W3113679918C187288502 @default.
- W3113679918 hasConceptScore W3113679918C204321447 @default.
- W3113679918 hasConceptScore W3113679918C2524010 @default.
- W3113679918 hasConceptScore W3113679918C25343380 @default.
- W3113679918 hasConceptScore W3113679918C2865642 @default.
- W3113679918 hasConceptScore W3113679918C33923547 @default.
- W3113679918 hasConceptScore W3113679918C36503486 @default.
- W3113679918 hasConceptScore W3113679918C41008148 @default.
- W3113679918 hasConceptScore W3113679918C41608201 @default.
- W3113679918 hasConceptScore W3113679918C43122875 @default.
- W3113679918 hasConceptScore W3113679918C77805123 @default.
- W3113679918 hasConceptScore W3113679918C80444323 @default.
- W3113679918 hasConceptScore W3113679918C90805587 @default.
- W3113679918 hasLocation W31136799181 @default.
- W3113679918 hasLocation W31136799182 @default.
- W3113679918 hasLocation W31136799183 @default.
- W3113679918 hasOpenAccess W3113679918 @default.
- W3113679918 hasPrimaryLocation W31136799181 @default.
- W3113679918 hasRelatedWork W1680231578 @default.
- W3113679918 hasRelatedWork W2102646808 @default.
- W3113679918 hasRelatedWork W2110891728 @default.
- W3113679918 hasRelatedWork W2151108588 @default.