FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783820753> ?p ?o ?g. }

• W2783820753 abstract "Humans can quickly learn new visual concepts, perhaps because they can easily visualize or imagine what novel objects look like from different views. Incorporating this ability to hallucinate novel instances of new concepts might help machine vision systems perform better low-shot learning, i.e., learning concepts from few examples. We present a novel approach to low-shot learning that uses this idea. Our approach builds on recent progress in meta-learning (learning to learn) by combining a meta-learner with a hallucinator that produces additional training examples, and optimizing both models jointly. Our hallucinator can be incorporated into a variety of meta-learners and provides significant gains: up to a 6 point boost in classification accuracy when only a single training example is available, yielding state-of-the-art performance on the challenging ImageNet low-shot classification benchmark." @default.
• W2783820753 created "2018-01-26" @default.
• W2783820753 creator A5036548506 @default.
• W2783820753 creator A5039231803 @default.
• W2783820753 creator A5049246408 @default.
• W2783820753 creator A5075246991 @default.
• W2783820753 date "2018-01-16" @default.
• W2783820753 modified "2023-10-16" @default.
• W2783820753 title "Low-Shot Learning from Imaginary Data" @default.
• W2783820753 cites W1151768060 @default.
• W2783820753 cites W1488195538 @default.
• W2783820753 cites W1800356822 @default.
• W2783820753 cites W1909320841 @default.
• W2783820753 cites W2099471712 @default.
• W2783820753 cites W2115733720 @default.
• W2783820753 cites W2118330267 @default.
• W2783820753 cites W2126204609 @default.
• W2783820753 cites W2133774033 @default.
• W2783820753 cites W2138621090 @default.
• W2783820753 cites W2144656844 @default.
• W2783820753 cites W2150385401 @default.
• W2783820753 cites W2155853132 @default.
• W2783820753 cites W2173520492 @default.
• W2783820753 cites W2187089797 @default.
• W2783820753 cites W2412589713 @default.
• W2783820753 cites W2432004435 @default.
• W2783820753 cites W2432717477 @default.
• W2783820753 cites W2435450765 @default.
• W2783820753 cites W2472819217 @default.
• W2783820753 cites W2519882289 @default.
• W2783820753 cites W2585027717 @default.
• W2783820753 cites W2589226201 @default.
• W2783820753 cites W2601450892 @default.
• W2783820753 cites W2619371851 @default.
• W2783820753 cites W2753160622 @default.
• W2783820753 cites W2753300133 @default.
• W2783820753 cites W2765332150 @default.
• W2783820753 cites W2949650786 @default.
• W2783820753 cites W2949823873 @default.
• W2783820753 cites W2951004968 @default.
• W2783820753 cites W2963078860 @default.
• W2783820753 cites W3091905774 @default.
• W2783820753 doi "https://doi.org/10.48550/arxiv.1801.05401" @default.
• W2783820753 hasPublicationYear "2018" @default.
• W2783820753 type Work @default.
• W2783820753 sameAs 2783820753 @default.
• W2783820753 citedByCount "52" @default.
• W2783820753 countsByYear W27838207532017 @default.
• W2783820753 countsByYear W27838207532018 @default.
• W2783820753 countsByYear W27838207532019 @default.
• W2783820753 countsByYear W27838207532020 @default.
• W2783820753 countsByYear W27838207532021 @default.
• W2783820753 crossrefType "posted-content" @default.
• W2783820753 hasAuthorship W2783820753A5036548506 @default.
• W2783820753 hasAuthorship W2783820753A5039231803 @default.
• W2783820753 hasAuthorship W2783820753A5049246408 @default.
• W2783820753 hasAuthorship W2783820753A5075246991 @default.
• W2783820753 hasBestOaLocation W27838207531 @default.
• W2783820753 hasConcept C119857082 @default.
• W2783820753 hasConcept C127413603 @default.
• W2783820753 hasConcept C13280743 @default.
• W2783820753 hasConcept C136197465 @default.
• W2783820753 hasConcept C154945302 @default.
• W2783820753 hasConcept C162324750 @default.
• W2783820753 hasConcept C178790620 @default.
• W2783820753 hasConcept C185592680 @default.
• W2783820753 hasConcept C185798385 @default.
• W2783820753 hasConcept C187736073 @default.
• W2783820753 hasConcept C205649164 @default.
• W2783820753 hasConcept C2524010 @default.
• W2783820753 hasConcept C2778344882 @default.
• W2783820753 hasConcept C2780451532 @default.
• W2783820753 hasConcept C2781002164 @default.
• W2783820753 hasConcept C28719098 @default.
• W2783820753 hasConcept C2911011789 @default.
• W2783820753 hasConcept C2992734406 @default.
• W2783820753 hasConcept C33923547 @default.
• W2783820753 hasConcept C41008148 @default.
• W2783820753 hasConcept C78519656 @default.
• W2783820753 hasConceptScore W2783820753C119857082 @default.
• W2783820753 hasConceptScore W2783820753C127413603 @default.
• W2783820753 hasConceptScore W2783820753C13280743 @default.
• W2783820753 hasConceptScore W2783820753C136197465 @default.
• W2783820753 hasConceptScore W2783820753C154945302 @default.
• W2783820753 hasConceptScore W2783820753C162324750 @default.
• W2783820753 hasConceptScore W2783820753C178790620 @default.
• W2783820753 hasConceptScore W2783820753C185592680 @default.
• W2783820753 hasConceptScore W2783820753C185798385 @default.
• W2783820753 hasConceptScore W2783820753C187736073 @default.
• W2783820753 hasConceptScore W2783820753C205649164 @default.
• W2783820753 hasConceptScore W2783820753C2524010 @default.
• W2783820753 hasConceptScore W2783820753C2778344882 @default.
• W2783820753 hasConceptScore W2783820753C2780451532 @default.
• W2783820753 hasConceptScore W2783820753C2781002164 @default.
• W2783820753 hasConceptScore W2783820753C28719098 @default.
• W2783820753 hasConceptScore W2783820753C2911011789 @default.
• W2783820753 hasConceptScore W2783820753C2992734406 @default.
• W2783820753 hasConceptScore W2783820753C33923547 @default.
• W2783820753 hasConceptScore W2783820753C41008148 @default.
• W2783820753 hasConceptScore W2783820753C78519656 @default.

• next