Matches in SemOpenAlex for { <https://semopenalex.org/work/W3185571403> ?p ?o ?g. }
- W3185571403 abstract "The goal of few-shot classification is to classify new categories with few labeled examples within each class. Nowadays, the excellent performance in handling few-shot classification problems is shown by metric-based meta-learning methods. However, it is very hard for previous methods to discriminate the fine-grained sub-categories in the embedding space without fine-grained labels. This may lead to unsatisfactory generalization to fine-grained subcategories, and thus affects model interpretation. To tackle this problem, we introduce the contrastive loss into few-shot classification for learning latent fine-grained structure in the embedding space. Furthermore, to overcome the drawbacks of random image transformation used in current contrastive learning in producing noisy and inaccurate image pairs (i.e., views), we develop a learning-to-learn algorithm to automatically generate different views of the same image. Extensive experiments on standard few-shot learning benchmarks demonstrate the superiority of our method." @default.
- W3185571403 created "2021-08-02" @default.
- W3185571403 creator A5004994916 @default.
- W3185571403 creator A5010510537 @default.
- W3185571403 creator A5015996294 @default.
- W3185571403 creator A5026752182 @default.
- W3185571403 creator A5051227924 @default.
- W3185571403 date "2021-07-19" @default.
- W3185571403 modified "2023-10-14" @default.
- W3185571403 title "Boosting Few-Shot Classification with View-Learnable Contrastive Learning" @default.
- W3185571403 cites W2115733720 @default.
- W3185571403 cites W2117539524 @default.
- W3185571403 cites W2138011018 @default.
- W3185571403 cites W2138621090 @default.
- W3185571403 cites W2601450892 @default.
- W3185571403 cites W2604763608 @default.
- W3185571403 cites W2625674597 @default.
- W3185571403 cites W2732026016 @default.
- W3185571403 cites W2787705367 @default.
- W3185571403 cites W2798991696 @default.
- W3185571403 cites W2904338933 @default.
- W3185571403 cites W2909383239 @default.
- W3185571403 cites W2917551568 @default.
- W3185571403 cites W2949517790 @default.
- W3185571403 cites W2962895018 @default.
- W3185571403 cites W2963341924 @default.
- W3185571403 cites W2963741406 @default.
- W3185571403 cites W2964105864 @default.
- W3185571403 cites W3005680577 @default.
- W3185571403 cites W3030509278 @default.
- W3185571403 cites W3035370595 @default.
- W3185571403 cites W3035524453 @default.
- W3185571403 cites W3097643695 @default.
- W3185571403 cites W3106428938 @default.
- W3185571403 cites W603908379 @default.
- W3185571403 cites W14333344 @default.
- W3185571403 doi "https://doi.org/10.48550/arxiv.2107.09242" @default.
- W3185571403 hasPublicationYear "2021" @default.
- W3185571403 type Work @default.
- W3185571403 sameAs 3185571403 @default.
- W3185571403 citedByCount "0" @default.
- W3185571403 crossrefType "posted-content" @default.
- W3185571403 hasAuthorship W3185571403A5004994916 @default.
- W3185571403 hasAuthorship W3185571403A5010510537 @default.
- W3185571403 hasAuthorship W3185571403A5015996294 @default.
- W3185571403 hasAuthorship W3185571403A5026752182 @default.
- W3185571403 hasAuthorship W3185571403A5051227924 @default.
- W3185571403 hasBestOaLocation W31855714031 @default.
- W3185571403 hasConcept C115961682 @default.
- W3185571403 hasConcept C119857082 @default.
- W3185571403 hasConcept C134306372 @default.
- W3185571403 hasConcept C153180895 @default.
- W3185571403 hasConcept C154945302 @default.
- W3185571403 hasConcept C162324750 @default.
- W3185571403 hasConcept C176217482 @default.
- W3185571403 hasConcept C177148314 @default.
- W3185571403 hasConcept C178790620 @default.
- W3185571403 hasConcept C185592680 @default.
- W3185571403 hasConcept C21547014 @default.
- W3185571403 hasConcept C2777212361 @default.
- W3185571403 hasConcept C2778344882 @default.
- W3185571403 hasConcept C33923547 @default.
- W3185571403 hasConcept C41008148 @default.
- W3185571403 hasConcept C41608201 @default.
- W3185571403 hasConcept C46686674 @default.
- W3185571403 hasConcept C75294576 @default.
- W3185571403 hasConceptScore W3185571403C115961682 @default.
- W3185571403 hasConceptScore W3185571403C119857082 @default.
- W3185571403 hasConceptScore W3185571403C134306372 @default.
- W3185571403 hasConceptScore W3185571403C153180895 @default.
- W3185571403 hasConceptScore W3185571403C154945302 @default.
- W3185571403 hasConceptScore W3185571403C162324750 @default.
- W3185571403 hasConceptScore W3185571403C176217482 @default.
- W3185571403 hasConceptScore W3185571403C177148314 @default.
- W3185571403 hasConceptScore W3185571403C178790620 @default.
- W3185571403 hasConceptScore W3185571403C185592680 @default.
- W3185571403 hasConceptScore W3185571403C21547014 @default.
- W3185571403 hasConceptScore W3185571403C2777212361 @default.
- W3185571403 hasConceptScore W3185571403C2778344882 @default.
- W3185571403 hasConceptScore W3185571403C33923547 @default.
- W3185571403 hasConceptScore W3185571403C41008148 @default.
- W3185571403 hasConceptScore W3185571403C41608201 @default.
- W3185571403 hasConceptScore W3185571403C46686674 @default.
- W3185571403 hasConceptScore W3185571403C75294576 @default.
- W3185571403 hasLocation W31855714031 @default.
- W3185571403 hasOpenAccess W3185571403 @default.
- W3185571403 hasPrimaryLocation W31855714031 @default.
- W3185571403 hasRelatedWork W1996541855 @default.
- W3185571403 hasRelatedWork W2003125512 @default.
- W3185571403 hasRelatedWork W2010832783 @default.
- W3185571403 hasRelatedWork W2026121273 @default.
- W3185571403 hasRelatedWork W2108043611 @default.
- W3185571403 hasRelatedWork W2361785652 @default.
- W3185571403 hasRelatedWork W2926953305 @default.
- W3185571403 hasRelatedWork W2944309180 @default.
- W3185571403 hasRelatedWork W2989932438 @default.
- W3185571403 hasRelatedWork W3095362205 @default.
- W3185571403 isParatext "false" @default.
- W3185571403 isRetracted "false" @default.
- W3185571403 magId "3185571403" @default.