Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288358251> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4288358251 abstract "In this work, we aim to realize a method for embedding human knowledge into deep neural networks. While the conventional method to embed human knowledge has been applied for non-deep machine learning, it is challenging to apply it for deep learning models due to the enormous number of model parameters. To tackle this problem, we focus on the attention mechanism of an attention branch network (ABN). In this paper, we propose a fine-tuning method that utilizes a single-channel attention map which is manually edited by a human expert. Our fine-tuning method can train a network so that the output attention map corresponds to the edited ones. As a result, the fine-tuned network can output an attention map that takes into account human knowledge. Experimental results with ImageNet, CUB-200-2010, and IDRiD demonstrate that it is possible to obtain a clear attention map for a visual explanation and improve the classification performance. Our findings can be a novel framework for optimizing networks through human intuitive editing via a visual interface and suggest new possibilities for human-machine cooperation in addition to the improvement of visual explanations." @default.
- W4288358251 created "2022-07-29" @default.
- W4288358251 creator A5011810133 @default.
- W4288358251 creator A5031685667 @default.
- W4288358251 creator A5053061807 @default.
- W4288358251 creator A5056234403 @default.
- W4288358251 creator A5062009349 @default.
- W4288358251 creator A5082397605 @default.
- W4288358251 creator A5089532441 @default.
- W4288358251 date "2019-05-09" @default.
- W4288358251 modified "2023-10-16" @default.
- W4288358251 title "Embedding Human Knowledge into Deep Neural Network via Attention Map" @default.
- W4288358251 doi "https://doi.org/10.48550/arxiv.1905.03540" @default.
- W4288358251 hasPublicationYear "2019" @default.
- W4288358251 type Work @default.
- W4288358251 citedByCount "2" @default.
- W4288358251 countsByYear W42883582512022 @default.
- W4288358251 countsByYear W42883582512023 @default.
- W4288358251 crossrefType "posted-content" @default.
- W4288358251 hasAuthorship W4288358251A5011810133 @default.
- W4288358251 hasAuthorship W4288358251A5031685667 @default.
- W4288358251 hasAuthorship W4288358251A5053061807 @default.
- W4288358251 hasAuthorship W4288358251A5056234403 @default.
- W4288358251 hasAuthorship W4288358251A5062009349 @default.
- W4288358251 hasAuthorship W4288358251A5082397605 @default.
- W4288358251 hasAuthorship W4288358251A5089532441 @default.
- W4288358251 hasBestOaLocation W42883582511 @default.
- W4288358251 hasConcept C108583219 @default.
- W4288358251 hasConcept C119857082 @default.
- W4288358251 hasConcept C120665830 @default.
- W4288358251 hasConcept C121332964 @default.
- W4288358251 hasConcept C154945302 @default.
- W4288358251 hasConcept C192209626 @default.
- W4288358251 hasConcept C2984842247 @default.
- W4288358251 hasConcept C41008148 @default.
- W4288358251 hasConcept C41608201 @default.
- W4288358251 hasConcept C50644808 @default.
- W4288358251 hasConceptScore W4288358251C108583219 @default.
- W4288358251 hasConceptScore W4288358251C119857082 @default.
- W4288358251 hasConceptScore W4288358251C120665830 @default.
- W4288358251 hasConceptScore W4288358251C121332964 @default.
- W4288358251 hasConceptScore W4288358251C154945302 @default.
- W4288358251 hasConceptScore W4288358251C192209626 @default.
- W4288358251 hasConceptScore W4288358251C2984842247 @default.
- W4288358251 hasConceptScore W4288358251C41008148 @default.
- W4288358251 hasConceptScore W4288358251C41608201 @default.
- W4288358251 hasConceptScore W4288358251C50644808 @default.
- W4288358251 hasLocation W42883582511 @default.
- W4288358251 hasOpenAccess W4288358251 @default.
- W4288358251 hasPrimaryLocation W42883582511 @default.
- W4288358251 hasRelatedWork W2604802223 @default.
- W4288358251 hasRelatedWork W2791691546 @default.
- W4288358251 hasRelatedWork W2950066684 @default.
- W4288358251 hasRelatedWork W2978367927 @default.
- W4288358251 hasRelatedWork W3034749675 @default.
- W4288358251 hasRelatedWork W3124304076 @default.
- W4288358251 hasRelatedWork W3162132941 @default.
- W4288358251 hasRelatedWork W3174422331 @default.
- W4288358251 hasRelatedWork W3179488938 @default.
- W4288358251 hasRelatedWork W4210508258 @default.
- W4288358251 isParatext "false" @default.
- W4288358251 isRetracted "false" @default.
- W4288358251 workType "article" @default.