Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963485453> ?p ?o ?g. }
- W2963485453 abstract "We introduce KBGAN, an adversarial learning framework to improve the performances of a wide range of existing knowledge graph embedding models. Because knowledge graphs typically only contain positive facts, sampling useful negative training examples is a nontrivial task. Replacing the head or tail entity of a fact with a uniformly randomly selected entity is a conventional method for generating negative facts, but the majority of the generated negative facts can be easily discriminated from positive facts, and will contribute little towards the training. Inspired by generative adversarial networks (GANs), we use one knowledge graph embedding model as a negative sample generator to assist the training of our desired model, which acts as the discriminator in GANs. This framework is independent of the concrete form of generator and discriminator, and therefore can utilize a wide variety of knowledge graph embedding models as its building blocks. In experiments, we adversarially train two translation-based models, TRANSE and TRANSD, each with assistance from one of the two probability-based models, DISTMULT and COMPLEX. We evaluate the performances of KBGAN on the link prediction task, using three knowledge base completion datasets: FB15k-237, WN18 and WN18RR. Experimental results show that adversarial training substantially improves the performances of target embedding models under various settings." @default.
- W2963485453 created "2019-07-30" @default.
- W2963485453 creator A5050195037 @default.
- W2963485453 creator A5088291592 @default.
- W2963485453 date "2018-01-01" @default.
- W2963485453 modified "2023-10-16" @default.
- W2963485453 title "KBGAN: Adversarial Learning for Knowledge Graph Embeddings" @default.
- W2963485453 cites W1529533208 @default.
- W2963485453 cites W1533230146 @default.
- W2963485453 cites W2016753842 @default.
- W2963485453 cites W2022166150 @default.
- W2963485453 cites W205829674 @default.
- W2963485453 cites W2094728533 @default.
- W2963485453 cites W2099471712 @default.
- W2963485453 cites W2125389028 @default.
- W2963485453 cites W2145544171 @default.
- W2963485453 cites W2155027007 @default.
- W2963485453 cites W2184957013 @default.
- W2963485453 cites W2226877337 @default.
- W2963485453 cites W2250184916 @default.
- W2963485453 cites W2250342289 @default.
- W2963485453 cites W2283196293 @default.
- W2963485453 cites W2607491080 @default.
- W2963485453 cites W2963432357 @default.
- W2963485453 cites W2963534707 @default.
- W2963485453 cites W2964116313 @default.
- W2963485453 cites W2964121744 @default.
- W2963485453 cites W2964268978 @default.
- W2963485453 cites W3101023724 @default.
- W2963485453 doi "https://doi.org/10.18653/v1/n18-1133" @default.
- W2963485453 hasPublicationYear "2018" @default.
- W2963485453 type Work @default.
- W2963485453 sameAs 2963485453 @default.
- W2963485453 citedByCount "165" @default.
- W2963485453 countsByYear W29634854532017 @default.
- W2963485453 countsByYear W29634854532018 @default.
- W2963485453 countsByYear W29634854532019 @default.
- W2963485453 countsByYear W29634854532020 @default.
- W2963485453 countsByYear W29634854532021 @default.
- W2963485453 countsByYear W29634854532022 @default.
- W2963485453 countsByYear W29634854532023 @default.
- W2963485453 crossrefType "proceedings-article" @default.
- W2963485453 hasAuthorship W2963485453A5050195037 @default.
- W2963485453 hasAuthorship W2963485453A5088291592 @default.
- W2963485453 hasBestOaLocation W29634854531 @default.
- W2963485453 hasConcept C119857082 @default.
- W2963485453 hasConcept C121332964 @default.
- W2963485453 hasConcept C132525143 @default.
- W2963485453 hasConcept C154945302 @default.
- W2963485453 hasConcept C162324750 @default.
- W2963485453 hasConcept C163258240 @default.
- W2963485453 hasConcept C187736073 @default.
- W2963485453 hasConcept C2779803651 @default.
- W2963485453 hasConcept C2780451532 @default.
- W2963485453 hasConcept C2780992000 @default.
- W2963485453 hasConcept C2987255567 @default.
- W2963485453 hasConcept C37736160 @default.
- W2963485453 hasConcept C39890363 @default.
- W2963485453 hasConcept C41008148 @default.
- W2963485453 hasConcept C41608201 @default.
- W2963485453 hasConcept C4554734 @default.
- W2963485453 hasConcept C62520636 @default.
- W2963485453 hasConcept C75564084 @default.
- W2963485453 hasConcept C76155785 @default.
- W2963485453 hasConcept C80444323 @default.
- W2963485453 hasConcept C94915269 @default.
- W2963485453 hasConceptScore W2963485453C119857082 @default.
- W2963485453 hasConceptScore W2963485453C121332964 @default.
- W2963485453 hasConceptScore W2963485453C132525143 @default.
- W2963485453 hasConceptScore W2963485453C154945302 @default.
- W2963485453 hasConceptScore W2963485453C162324750 @default.
- W2963485453 hasConceptScore W2963485453C163258240 @default.
- W2963485453 hasConceptScore W2963485453C187736073 @default.
- W2963485453 hasConceptScore W2963485453C2779803651 @default.
- W2963485453 hasConceptScore W2963485453C2780451532 @default.
- W2963485453 hasConceptScore W2963485453C2780992000 @default.
- W2963485453 hasConceptScore W2963485453C2987255567 @default.
- W2963485453 hasConceptScore W2963485453C37736160 @default.
- W2963485453 hasConceptScore W2963485453C39890363 @default.
- W2963485453 hasConceptScore W2963485453C41008148 @default.
- W2963485453 hasConceptScore W2963485453C41608201 @default.
- W2963485453 hasConceptScore W2963485453C4554734 @default.
- W2963485453 hasConceptScore W2963485453C62520636 @default.
- W2963485453 hasConceptScore W2963485453C75564084 @default.
- W2963485453 hasConceptScore W2963485453C76155785 @default.
- W2963485453 hasConceptScore W2963485453C80444323 @default.
- W2963485453 hasConceptScore W2963485453C94915269 @default.
- W2963485453 hasLocation W29634854531 @default.
- W2963485453 hasLocation W29634854532 @default.
- W2963485453 hasOpenAccess W2963485453 @default.
- W2963485453 hasPrimaryLocation W29634854531 @default.
- W2963485453 hasRelatedWork W1524858705 @default.
- W2963485453 hasRelatedWork W2768762802 @default.
- W2963485453 hasRelatedWork W2923818335 @default.
- W2963485453 hasRelatedWork W2945796017 @default.
- W2963485453 hasRelatedWork W2953770707 @default.
- W2963485453 hasRelatedWork W2964583308 @default.
- W2963485453 hasRelatedWork W3029957763 @default.
- W2963485453 hasRelatedWork W3034334914 @default.
- W2963485453 hasRelatedWork W4213074484 @default.