Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381245780> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4381245780 abstract "Despite the advances made in visual object recognition, state-of-the-art deep learning models struggle to effectively recognize novel objects in a few-shot setting where only a limited number of examples are provided. Unlike humans who excel at such tasks, these models often fail to leverage known relationships between entities in order to draw conclusions about such objects. In this work, we show that incorporating a symbolic knowledge graph into a state-of-the-art recognition model enables a new approach for effective few-shot classification. In our proposed neuro-symbolic architecture and training methodology, the knowledge graph is augmented with additional relationships extracted from a small set of examples, improving its ability to recognize novel objects by considering the presence of interconnected entities. Unlike existing few-shot classifiers, we show that this enables our model to incorporate not only objects but also abstract concepts and affordances. The existence of the knowledge graph also makes this approach amenable to interpretability through analysis of the relationships contained within it. We empirically show that our approach outperforms current state-of-the-art few-shot multi-label classification methods on the COCO dataset and evaluate the addition of abstract concepts and affordances on the Visual Genome dataset." @default.
- W4381245780 created "2023-06-20" @default.
- W4381245780 creator A5048453008 @default.
- W4381245780 creator A5048668442 @default.
- W4381245780 creator A5079864567 @default.
- W4381245780 creator A5087505541 @default.
- W4381245780 date "2023-06-15" @default.
- W4381245780 modified "2023-09-23" @default.
- W4381245780 title "Sample-Efficient Learning of Novel Visual Concepts" @default.
- W4381245780 doi "https://doi.org/10.48550/arxiv.2306.09482" @default.
- W4381245780 hasPublicationYear "2023" @default.
- W4381245780 type Work @default.
- W4381245780 citedByCount "0" @default.
- W4381245780 crossrefType "posted-content" @default.
- W4381245780 hasAuthorship W4381245780A5048453008 @default.
- W4381245780 hasAuthorship W4381245780A5048668442 @default.
- W4381245780 hasAuthorship W4381245780A5079864567 @default.
- W4381245780 hasAuthorship W4381245780A5087505541 @default.
- W4381245780 hasBestOaLocation W43812457801 @default.
- W4381245780 hasConcept C107457646 @default.
- W4381245780 hasConcept C119857082 @default.
- W4381245780 hasConcept C132525143 @default.
- W4381245780 hasConcept C153083717 @default.
- W4381245780 hasConcept C154945302 @default.
- W4381245780 hasConcept C177264268 @default.
- W4381245780 hasConcept C194995250 @default.
- W4381245780 hasConcept C199360897 @default.
- W4381245780 hasConcept C2781067378 @default.
- W4381245780 hasConcept C41008148 @default.
- W4381245780 hasConcept C80444323 @default.
- W4381245780 hasConceptScore W4381245780C107457646 @default.
- W4381245780 hasConceptScore W4381245780C119857082 @default.
- W4381245780 hasConceptScore W4381245780C132525143 @default.
- W4381245780 hasConceptScore W4381245780C153083717 @default.
- W4381245780 hasConceptScore W4381245780C154945302 @default.
- W4381245780 hasConceptScore W4381245780C177264268 @default.
- W4381245780 hasConceptScore W4381245780C194995250 @default.
- W4381245780 hasConceptScore W4381245780C199360897 @default.
- W4381245780 hasConceptScore W4381245780C2781067378 @default.
- W4381245780 hasConceptScore W4381245780C41008148 @default.
- W4381245780 hasConceptScore W4381245780C80444323 @default.
- W4381245780 hasLocation W43812457801 @default.
- W4381245780 hasOpenAccess W4381245780 @default.
- W4381245780 hasPrimaryLocation W43812457801 @default.
- W4381245780 hasRelatedWork W2605281151 @default.
- W4381245780 hasRelatedWork W3006943036 @default.
- W4381245780 hasRelatedWork W3012234327 @default.
- W4381245780 hasRelatedWork W3023163568 @default.
- W4381245780 hasRelatedWork W3119715496 @default.
- W4381245780 hasRelatedWork W3191046242 @default.
- W4381245780 hasRelatedWork W4205364923 @default.
- W4381245780 hasRelatedWork W4206534706 @default.
- W4381245780 hasRelatedWork W4229079080 @default.
- W4381245780 hasRelatedWork W4294031299 @default.
- W4381245780 isParatext "false" @default.
- W4381245780 isRetracted "false" @default.
- W4381245780 workType "article" @default.