Matches in SemOpenAlex for { <https://semopenalex.org/work/W3039442841> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3039442841 abstract "Visual Question Answering(VQA) is a highly complex problem set, relying on many sub-problems to produce reasonable answers. In this paper, we present the hypothesis that Visual Question Answering should be viewed as a multi-task problem, and provide evidence to support this hypothesis. We demonstrate this by reformatting two commonly used Visual Question Answering datasets, COCO-QA and DAQUAR, into a multi-task format and train these reformatted datasets on two baseline networks, with one designed specifically to eliminate other possible causes for performance changes as a result of the reformatting. Though the networks demonstrated in this paper do not achieve strongly competitive results, we find that the multi-task approach to Visual Question Answering results in increases in performance of 5-9% against the single-task formatting, and that the networks reach convergence much faster than in the single-task case. Finally we discuss possible reasons for the observed difference in performance, and perform additional experiments which rule out causes not associated with the learning of the dataset as a multi-task problem." @default.
- W3039442841 created "2020-07-10" @default.
- W3039442841 creator A5037018261 @default.
- W3039442841 creator A5064060276 @default.
- W3039442841 date "2020-07-03" @default.
- W3039442841 modified "2023-09-27" @default.
- W3039442841 title "Visual Question Answering as a Multi-Task Problem." @default.
- W3039442841 cites W1575833922 @default.
- W3039442841 cites W1686810756 @default.
- W3039442841 cites W1895641373 @default.
- W3039442841 cites W1933349210 @default.
- W3039442841 cites W2060277733 @default.
- W3039442841 cites W2142192571 @default.
- W3039442841 cites W2151498684 @default.
- W3039442841 cites W2170738476 @default.
- W3039442841 cites W2250539671 @default.
- W3039442841 cites W2346425926 @default.
- W3039442841 cites W2766860085 @default.
- W3039442841 cites W2776855315 @default.
- W3039442841 cites W2903382683 @default.
- W3039442841 cites W2913340405 @default.
- W3039442841 cites W2962749469 @default.
- W3039442841 cites W2963047717 @default.
- W3039442841 cites W2963622213 @default.
- W3039442841 cites W2963656855 @default.
- W3039442841 cites W2963672682 @default.
- W3039442841 cites W2963890019 @default.
- W3039442841 cites W2963954913 @default.
- W3039442841 cites W627986001 @default.
- W3039442841 hasPublicationYear "2020" @default.
- W3039442841 type Work @default.
- W3039442841 sameAs 3039442841 @default.
- W3039442841 citedByCount "1" @default.
- W3039442841 countsByYear W30394428412020 @default.
- W3039442841 crossrefType "posted-content" @default.
- W3039442841 hasAuthorship W3039442841A5037018261 @default.
- W3039442841 hasAuthorship W3039442841A5064060276 @default.
- W3039442841 hasConcept C111368507 @default.
- W3039442841 hasConcept C111919701 @default.
- W3039442841 hasConcept C119857082 @default.
- W3039442841 hasConcept C12725497 @default.
- W3039442841 hasConcept C127313418 @default.
- W3039442841 hasConcept C154945302 @default.
- W3039442841 hasConcept C162324750 @default.
- W3039442841 hasConcept C177264268 @default.
- W3039442841 hasConcept C187736073 @default.
- W3039442841 hasConcept C199360897 @default.
- W3039442841 hasConcept C23123220 @default.
- W3039442841 hasConcept C2780451532 @default.
- W3039442841 hasConcept C41008148 @default.
- W3039442841 hasConcept C44291984 @default.
- W3039442841 hasConcept C88006597 @default.
- W3039442841 hasConceptScore W3039442841C111368507 @default.
- W3039442841 hasConceptScore W3039442841C111919701 @default.
- W3039442841 hasConceptScore W3039442841C119857082 @default.
- W3039442841 hasConceptScore W3039442841C12725497 @default.
- W3039442841 hasConceptScore W3039442841C127313418 @default.
- W3039442841 hasConceptScore W3039442841C154945302 @default.
- W3039442841 hasConceptScore W3039442841C162324750 @default.
- W3039442841 hasConceptScore W3039442841C177264268 @default.
- W3039442841 hasConceptScore W3039442841C187736073 @default.
- W3039442841 hasConceptScore W3039442841C199360897 @default.
- W3039442841 hasConceptScore W3039442841C23123220 @default.
- W3039442841 hasConceptScore W3039442841C2780451532 @default.
- W3039442841 hasConceptScore W3039442841C41008148 @default.
- W3039442841 hasConceptScore W3039442841C44291984 @default.
- W3039442841 hasConceptScore W3039442841C88006597 @default.
- W3039442841 hasLocation W30394428411 @default.
- W3039442841 hasOpenAccess W3039442841 @default.
- W3039442841 hasPrimaryLocation W30394428411 @default.
- W3039442841 hasRelatedWork W2339712187 @default.
- W3039442841 hasRelatedWork W2769477337 @default.
- W3039442841 hasRelatedWork W2794284373 @default.
- W3039442841 hasRelatedWork W2903184186 @default.
- W3039442841 hasRelatedWork W2947962063 @default.
- W3039442841 hasRelatedWork W2951261952 @default.
- W3039442841 hasRelatedWork W2962781483 @default.
- W3039442841 hasRelatedWork W2963682631 @default.
- W3039442841 hasRelatedWork W2963976294 @default.
- W3039442841 hasRelatedWork W2970019270 @default.
- W3039442841 hasRelatedWork W2970229707 @default.
- W3039442841 hasRelatedWork W2980847631 @default.
- W3039442841 hasRelatedWork W2982389186 @default.
- W3039442841 hasRelatedWork W2982571117 @default.
- W3039442841 hasRelatedWork W2997399961 @default.
- W3039442841 hasRelatedWork W2997561853 @default.
- W3039442841 hasRelatedWork W3098382480 @default.
- W3039442841 hasRelatedWork W3159938639 @default.
- W3039442841 hasRelatedWork W3198817488 @default.
- W3039442841 hasRelatedWork W3207889266 @default.
- W3039442841 isParatext "false" @default.
- W3039442841 isRetracted "false" @default.
- W3039442841 magId "3039442841" @default.
- W3039442841 workType "article" @default.