Matches in SemOpenAlex for { <https://semopenalex.org/work/W2786772157> ?p ?o ?g. }
- W2786772157 abstract "Learning a deep model from small data is yet an opening and challenging problem. We focus on one-shot classification by deep learning approach based on a small quantity of training samples. We proposed a novel deep learning approach named Local Contrast Learning (LCL) based on the key insight about a human cognitive behavior that human recognizes the objects in a specific context by contrasting the objects in the context or in her/his memory. LCL is used to train a deep model that can contrast the recognizing sample with a couple of contrastive samples randomly drawn and shuffled. On one-shot classification task on Omniglot, the deep model based LCL with 122 layers and 1.94 millions of parameters, which was trained on a tiny dataset with only 60 classes and 20 samples per class, achieved the accuracy 97.99% that outperforms human and state-of-the-art established by Bayesian Program Learning (BPL) trained on 964 classes. LCL is a fundamental idea which can be applied to alleviate parametric model's overfitting resulted by lack of training samples." @default.
- W2786772157 created "2018-02-23" @default.
- W2786772157 creator A5002008893 @default.
- W2786772157 creator A5009262956 @default.
- W2786772157 creator A5022802322 @default.
- W2786772157 creator A5038878717 @default.
- W2786772157 creator A5044238570 @default.
- W2786772157 creator A5066223986 @default.
- W2786772157 date "2018-02-10" @default.
- W2786772157 modified "2023-09-26" @default.
- W2786772157 title "Local Contrast Learning." @default.
- W2786772157 cites W104184427 @default.
- W2786772157 cites W1498436455 @default.
- W2786772157 cites W1677182931 @default.
- W2786772157 cites W1799366690 @default.
- W2786772157 cites W1984314602 @default.
- W2786772157 cites W2114168642 @default.
- W2786772157 cites W2119305530 @default.
- W2786772157 cites W2138621090 @default.
- W2786772157 cites W2194321275 @default.
- W2786772157 cites W2321691871 @default.
- W2786772157 cites W2412589713 @default.
- W2786772157 cites W2472819217 @default.
- W2786772157 cites W2919115771 @default.
- W2786772157 cites W2949117887 @default.
- W2786772157 cites W2949427019 @default.
- W2786772157 cites W2949650786 @default.
- W2786772157 cites W2950537964 @default.
- W2786772157 cites W2953044442 @default.
- W2786772157 cites W2963341924 @default.
- W2786772157 cites W2963713828 @default.
- W2786772157 cites W2963775347 @default.
- W2786772157 cites W3037856073 @default.
- W2786772157 cites W3091905774 @default.
- W2786772157 hasPublicationYear "2018" @default.
- W2786772157 type Work @default.
- W2786772157 sameAs 2786772157 @default.
- W2786772157 citedByCount "0" @default.
- W2786772157 crossrefType "posted-content" @default.
- W2786772157 hasAuthorship W2786772157A5002008893 @default.
- W2786772157 hasAuthorship W2786772157A5009262956 @default.
- W2786772157 hasAuthorship W2786772157A5022802322 @default.
- W2786772157 hasAuthorship W2786772157A5038878717 @default.
- W2786772157 hasAuthorship W2786772157A5044238570 @default.
- W2786772157 hasAuthorship W2786772157A5066223986 @default.
- W2786772157 hasConcept C105795698 @default.
- W2786772157 hasConcept C107673813 @default.
- W2786772157 hasConcept C108583219 @default.
- W2786772157 hasConcept C117251300 @default.
- W2786772157 hasConcept C119857082 @default.
- W2786772157 hasConcept C120665830 @default.
- W2786772157 hasConcept C121332964 @default.
- W2786772157 hasConcept C153180895 @default.
- W2786772157 hasConcept C154945302 @default.
- W2786772157 hasConcept C162324750 @default.
- W2786772157 hasConcept C166957645 @default.
- W2786772157 hasConcept C185592680 @default.
- W2786772157 hasConcept C187736073 @default.
- W2786772157 hasConcept C192209626 @default.
- W2786772157 hasConcept C198531522 @default.
- W2786772157 hasConcept C205649164 @default.
- W2786772157 hasConcept C22019652 @default.
- W2786772157 hasConcept C2776502983 @default.
- W2786772157 hasConcept C2777212361 @default.
- W2786772157 hasConcept C2779343474 @default.
- W2786772157 hasConcept C2780451532 @default.
- W2786772157 hasConcept C33923547 @default.
- W2786772157 hasConcept C41008148 @default.
- W2786772157 hasConcept C43617362 @default.
- W2786772157 hasConcept C50644808 @default.
- W2786772157 hasConceptScore W2786772157C105795698 @default.
- W2786772157 hasConceptScore W2786772157C107673813 @default.
- W2786772157 hasConceptScore W2786772157C108583219 @default.
- W2786772157 hasConceptScore W2786772157C117251300 @default.
- W2786772157 hasConceptScore W2786772157C119857082 @default.
- W2786772157 hasConceptScore W2786772157C120665830 @default.
- W2786772157 hasConceptScore W2786772157C121332964 @default.
- W2786772157 hasConceptScore W2786772157C153180895 @default.
- W2786772157 hasConceptScore W2786772157C154945302 @default.
- W2786772157 hasConceptScore W2786772157C162324750 @default.
- W2786772157 hasConceptScore W2786772157C166957645 @default.
- W2786772157 hasConceptScore W2786772157C185592680 @default.
- W2786772157 hasConceptScore W2786772157C187736073 @default.
- W2786772157 hasConceptScore W2786772157C192209626 @default.
- W2786772157 hasConceptScore W2786772157C198531522 @default.
- W2786772157 hasConceptScore W2786772157C205649164 @default.
- W2786772157 hasConceptScore W2786772157C22019652 @default.
- W2786772157 hasConceptScore W2786772157C2776502983 @default.
- W2786772157 hasConceptScore W2786772157C2777212361 @default.
- W2786772157 hasConceptScore W2786772157C2779343474 @default.
- W2786772157 hasConceptScore W2786772157C2780451532 @default.
- W2786772157 hasConceptScore W2786772157C33923547 @default.
- W2786772157 hasConceptScore W2786772157C41008148 @default.
- W2786772157 hasConceptScore W2786772157C43617362 @default.
- W2786772157 hasConceptScore W2786772157C50644808 @default.
- W2786772157 hasLocation W27867721571 @default.
- W2786772157 hasOpenAccess W2786772157 @default.
- W2786772157 hasPrimaryLocation W27867721571 @default.
- W2786772157 hasRelatedWork W2465668693 @default.
- W2786772157 hasRelatedWork W2766914988 @default.