Matches in SemOpenAlex for { <https://semopenalex.org/work/W3202392894> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W3202392894 endingPage "373" @default.
- W3202392894 startingPage "357" @default.
- W3202392894 abstract "Traditional computer vision approaches, based on neural networks (NN), are typically trained on a large amount of image data. By minimizing the cross-entropy loss between a prediction and a given class label, the NN and its visual embedding space are learned to fulfill a given task. However, due to the sole dependence on the image data distribution of the training domain, these models tend to fail when applied to a target domain that differs from their source domain. To learn a more robust NN to domain shifts, we propose the knowledge graph neural network (KG-NN), a neuro-symbolic approach that supervises the training using image-data-invariant auxiliary knowledge. The auxiliary knowledge is first encoded in a knowledge graph with respective concepts and their relationships, which is then transformed into a dense vector representation via an embedding method. Using a contrastive loss function, KG-NN learns to adapt its visual embedding space and thus its weights according to the image-data invariant knowledge graph embedding space. We evaluate KG-NN on visual transfer learning tasks for classification using the mini-ImageNet dataset and its derivatives, as well as road sign recognition datasets from Germany and China. The results show that a visual model trained with a knowledge graph as a trainer outperforms a model trained with cross-entropy in all experiments, in particular when the domain gap increases. Besides better performance and stronger robustness to domain shifts, these KG-NN adapts to multiple datasets and classes without suffering heavily from catastrophic forgetting." @default.
- W3202392894 created "2021-10-11" @default.
- W3202392894 creator A5000758128 @default.
- W3202392894 creator A5031926202 @default.
- W3202392894 creator A5042442907 @default.
- W3202392894 creator A5065068289 @default.
- W3202392894 date "2021-01-01" @default.
- W3202392894 modified "2023-10-17" @default.
- W3202392894 title "Learning Visual Models Using a Knowledge Graph as a Trainer" @default.
- W3202392894 cites W102708294 @default.
- W3202392894 cites W1480376833 @default.
- W3202392894 cites W2067713319 @default.
- W3202392894 cites W2081580037 @default.
- W3202392894 cites W2100031962 @default.
- W3202392894 cites W2117539524 @default.
- W3202392894 cites W2168217710 @default.
- W3202392894 cites W2171061940 @default.
- W3202392894 cites W2194775991 @default.
- W3202392894 cites W2250539671 @default.
- W3202392894 cites W2472350142 @default.
- W3202392894 cites W2561529111 @default.
- W3202392894 cites W2728059831 @default.
- W3202392894 cites W2774837955 @default.
- W3202392894 cites W2887280559 @default.
- W3202392894 cites W2904378456 @default.
- W3202392894 cites W2932399282 @default.
- W3202392894 cites W2949972983 @default.
- W3202392894 cites W2963052338 @default.
- W3202392894 cites W2963477107 @default.
- W3202392894 cites W2963486920 @default.
- W3202392894 cites W2964086552 @default.
- W3202392894 cites W2964094751 @default.
- W3202392894 cites W2964125718 @default.
- W3202392894 cites W3035524453 @default.
- W3202392894 cites W3035605097 @default.
- W3202392894 cites W3037492894 @default.
- W3202392894 cites W3108975329 @default.
- W3202392894 doi "https://doi.org/10.1007/978-3-030-88361-4_21" @default.
- W3202392894 hasPublicationYear "2021" @default.
- W3202392894 type Work @default.
- W3202392894 sameAs 3202392894 @default.
- W3202392894 citedByCount "5" @default.
- W3202392894 countsByYear W32023928942022 @default.
- W3202392894 countsByYear W32023928942023 @default.
- W3202392894 crossrefType "book-chapter" @default.
- W3202392894 hasAuthorship W3202392894A5000758128 @default.
- W3202392894 hasAuthorship W3202392894A5031926202 @default.
- W3202392894 hasAuthorship W3202392894A5042442907 @default.
- W3202392894 hasAuthorship W3202392894A5065068289 @default.
- W3202392894 hasBestOaLocation W32023928942 @default.
- W3202392894 hasConcept C119857082 @default.
- W3202392894 hasConcept C132525143 @default.
- W3202392894 hasConcept C138885662 @default.
- W3202392894 hasConcept C153180895 @default.
- W3202392894 hasConcept C154945302 @default.
- W3202392894 hasConcept C207685749 @default.
- W3202392894 hasConcept C41008148 @default.
- W3202392894 hasConcept C41608201 @default.
- W3202392894 hasConcept C41895202 @default.
- W3202392894 hasConcept C50644808 @default.
- W3202392894 hasConcept C7149132 @default.
- W3202392894 hasConcept C80444323 @default.
- W3202392894 hasConcept C81363708 @default.
- W3202392894 hasConceptScore W3202392894C119857082 @default.
- W3202392894 hasConceptScore W3202392894C132525143 @default.
- W3202392894 hasConceptScore W3202392894C138885662 @default.
- W3202392894 hasConceptScore W3202392894C153180895 @default.
- W3202392894 hasConceptScore W3202392894C154945302 @default.
- W3202392894 hasConceptScore W3202392894C207685749 @default.
- W3202392894 hasConceptScore W3202392894C41008148 @default.
- W3202392894 hasConceptScore W3202392894C41608201 @default.
- W3202392894 hasConceptScore W3202392894C41895202 @default.
- W3202392894 hasConceptScore W3202392894C50644808 @default.
- W3202392894 hasConceptScore W3202392894C7149132 @default.
- W3202392894 hasConceptScore W3202392894C80444323 @default.
- W3202392894 hasConceptScore W3202392894C81363708 @default.
- W3202392894 hasLocation W32023928941 @default.
- W3202392894 hasLocation W32023928942 @default.
- W3202392894 hasOpenAccess W3202392894 @default.
- W3202392894 hasPrimaryLocation W32023928941 @default.
- W3202392894 hasRelatedWork W2175746458 @default.
- W3202392894 hasRelatedWork W2732542196 @default.
- W3202392894 hasRelatedWork W2738221750 @default.
- W3202392894 hasRelatedWork W2760085659 @default.
- W3202392894 hasRelatedWork W2767651786 @default.
- W3202392894 hasRelatedWork W2883200793 @default.
- W3202392894 hasRelatedWork W3012978760 @default.
- W3202392894 hasRelatedWork W3027997911 @default.
- W3202392894 hasRelatedWork W3093612317 @default.
- W3202392894 hasRelatedWork W4287776258 @default.
- W3202392894 isParatext "false" @default.
- W3202392894 isRetracted "false" @default.
- W3202392894 magId "3202392894" @default.
- W3202392894 workType "book-chapter" @default.