FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288889641> ?p ?o ?g. }

• W4288889641 endingPage "153" @default.
• W4288889641 startingPage "146" @default.
• W4288889641 abstract "Object recognition and viewpoint estimation lie at the heart of visual understanding. Recent studies have suggested that convolutional neural networks (CNNs) fail to generalize to out-of-distribution (OOD) category–viewpoint combinations, that is, combinations not seen during training. Here we investigate when and how such OOD generalization may be possible by evaluating CNNs trained to classify both object category and three-dimensional viewpoint on OOD combinations, and identifying the neural mechanisms that facilitate such OOD generalization. We show that increasing the number of in-distribution combinations (data diversity) substantially improves generalization to OOD combinations, even with the same amount of training data. We compare learning category and viewpoint in separate and shared network architectures, and observe starkly different trends on in-distribution and OOD combinations, that is, while shared networks are helpful in distribution, separate networks significantly outperform shared ones at OOD combinations. Finally, we demonstrate that such OOD generalization is facilitated by the neural mechanism of specialization, that is, the emergence of two types of neuron—neurons selective to category and invariant to viewpoint, and vice versa. The combination of object recognition and viewpoint estimation is essential for visual understanding. However, convolutional neural networks often fail to generalize to object category–viewpoint combinations that were not seen during training. The authors investigate the impact of data diversity and architectural choices on the capability of generalizing to out-of-distribution combinations." @default.
• W4288889641 created "2022-07-31" @default.
• W4288889641 creator A5020830959 @default.
• W4288889641 creator A5024274700 @default.
• W4288889641 creator A5040427850 @default.
• W4288889641 creator A5042578984 @default.
• W4288889641 creator A5043151044 @default.
• W4288889641 creator A5051000016 @default.
• W4288889641 creator A5055261310 @default.
• W4288889641 creator A5070176712 @default.
• W4288889641 creator A5075155488 @default.
• W4288889641 date "2022-02-21" @default.
• W4288889641 modified "2023-10-11" @default.
• W4288889641 title "When and how convolutional neural networks generalize to out-of-distribution category–viewpoint combinations" @default.
• W4288889641 cites W1575599996 @default.
• W4288889641 cites W1591870335 @default.
• W4288889641 cites W1849277567 @default.
• W4288889641 cites W1949483711 @default.
• W4288889641 cites W1990037101 @default.
• W4288889641 cites W1991264156 @default.
• W4288889641 cites W1994197834 @default.
• W4288889641 cites W1994488211 @default.
• W4288889641 cites W2008839571 @default.
• W4288889641 cites W2031342017 @default.
• W4288889641 cites W2068777106 @default.
• W4288889641 cites W2101295242 @default.
• W4288889641 cites W2112796928 @default.
• W4288889641 cites W2115441154 @default.
• W4288889641 cites W2123456673 @default.
• W4288889641 cites W2128942651 @default.
• W4288889641 cites W2138011018 @default.
• W4288889641 cites W2183341477 @default.
• W4288889641 cites W2194775991 @default.
• W4288889641 cites W2471821491 @default.
• W4288889641 cites W2518803647 @default.
• W4288889641 cites W2549139847 @default.
• W4288889641 cites W2563100679 @default.
• W4288889641 cites W2793987571 @default.
• W4288889641 cites W2795697078 @default.
• W4288889641 cites W2897124911 @default.
• W4288889641 cites W2908124316 @default.
• W4288889641 cites W2913340405 @default.
• W4288889641 cites W2962748819 @default.
• W4288889641 cites W2963188159 @default.
• W4288889641 cites W2963446712 @default.
• W4288889641 cites W2963665806 @default.
• W4288889641 cites W2963749936 @default.
• W4288889641 cites W2963826402 @default.
• W4288889641 cites W2964056579 @default.
• W4288889641 cites W2964137095 @default.
• W4288889641 cites W2964239644 @default.
• W4288889641 cites W2968993450 @default.
• W4288889641 cites W2998545307 @default.
• W4288889641 cites W3035574168 @default.
• W4288889641 cites W3085093108 @default.
• W4288889641 cites W4206410067 @default.
• W4288889641 cites W4244723818 @default.
• W4288889641 doi "https://doi.org/10.1038/s42256-021-00437-5" @default.
• W4288889641 hasPublicationYear "2022" @default.
• W4288889641 type Work @default.
• W4288889641 citedByCount "5" @default.
• W4288889641 countsByYear W42888896412022 @default.
• W4288889641 countsByYear W42888896412023 @default.
• W4288889641 crossrefType "journal-article" @default.
• W4288889641 hasAuthorship W4288889641A5020830959 @default.
• W4288889641 hasAuthorship W4288889641A5024274700 @default.
• W4288889641 hasAuthorship W4288889641A5040427850 @default.
• W4288889641 hasAuthorship W4288889641A5042578984 @default.
• W4288889641 hasAuthorship W4288889641A5043151044 @default.
• W4288889641 hasAuthorship W4288889641A5051000016 @default.
• W4288889641 hasAuthorship W4288889641A5055261310 @default.
• W4288889641 hasAuthorship W4288889641A5070176712 @default.
• W4288889641 hasAuthorship W4288889641A5075155488 @default.
• W4288889641 hasBestOaLocation W42888896412 @default.
• W4288889641 hasConcept C119857082 @default.
• W4288889641 hasConcept C134306372 @default.
• W4288889641 hasConcept C153180895 @default.
• W4288889641 hasConcept C154945302 @default.
• W4288889641 hasConcept C177148314 @default.
• W4288889641 hasConcept C190470478 @default.
• W4288889641 hasConcept C2781238097 @default.
• W4288889641 hasConcept C33923547 @default.
• W4288889641 hasConcept C37914503 @default.
• W4288889641 hasConcept C41008148 @default.
• W4288889641 hasConcept C50644808 @default.
• W4288889641 hasConcept C64876066 @default.
• W4288889641 hasConcept C81363708 @default.
• W4288889641 hasConcept C94124525 @default.
• W4288889641 hasConceptScore W4288889641C119857082 @default.
• W4288889641 hasConceptScore W4288889641C134306372 @default.
• W4288889641 hasConceptScore W4288889641C153180895 @default.
• W4288889641 hasConceptScore W4288889641C154945302 @default.
• W4288889641 hasConceptScore W4288889641C177148314 @default.
• W4288889641 hasConceptScore W4288889641C190470478 @default.
• W4288889641 hasConceptScore W4288889641C2781238097 @default.
• W4288889641 hasConceptScore W4288889641C33923547 @default.
• W4288889641 hasConceptScore W4288889641C37914503 @default.
• W4288889641 hasConceptScore W4288889641C41008148 @default.

• next