Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204676188> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3204676188 abstract "Learning to hallucinate additional examples has recently been shown as a promising direction to address few-shot learning tasks. This work investigates two important yet overlooked natural supervision signals for guiding the hallucination process – (i) extrinsic: classifiers trained on hallucinated examples should be close to strong classifiers that would be learned from a large amount of real examples; and (ii) intrinsic: clusters of hallucinated and real examples belonging to the same class should be pulled together, while simultaneously pushing apart clusters of hallucinated and real examples from different classes. We achieve (i) by introducing an additional mentor model on data-abundant base classes for directing the hallucinator, and achieve (ii) by performing contrastive learning between hallucinated and real examples. As a general, model-agnostic framework, our dual mentor-and self-directed (DMAS) hallucinator significantly improves few-shot learning performance on widely-used benchmarks in various scenarios." @default.
- W3204676188 created "2021-10-11" @default.
- W3204676188 creator A5033229022 @default.
- W3204676188 creator A5037727565 @default.
- W3204676188 creator A5039231803 @default.
- W3204676188 creator A5062546146 @default.
- W3204676188 creator A5071983998 @default.
- W3204676188 creator A5075246991 @default.
- W3204676188 date "2021-10-01" @default.
- W3204676188 modified "2023-10-03" @default.
- W3204676188 title "Learning to Hallucinate Examples from Extrinsic and Intrinsic Supervision" @default.
- W3204676188 cites W2115733720 @default.
- W3204676188 cites W2294370754 @default.
- W3204676188 cites W2296073425 @default.
- W3204676188 cites W2473930607 @default.
- W3204676188 cites W2625674597 @default.
- W3204676188 cites W2798836702 @default.
- W3204676188 cites W2798991696 @default.
- W3204676188 cites W2799404913 @default.
- W3204676188 cites W2904218366 @default.
- W3204676188 cites W2909383239 @default.
- W3204676188 cites W2924476266 @default.
- W3204676188 cites W2943605315 @default.
- W3204676188 cites W2949879676 @default.
- W3204676188 cites W2962843773 @default.
- W3204676188 cites W2962895018 @default.
- W3204676188 cites W2962987395 @default.
- W3204676188 cites W2963070905 @default.
- W3204676188 cites W2963350370 @default.
- W3204676188 cites W2963420272 @default.
- W3204676188 cites W2963845150 @default.
- W3204676188 cites W2963943197 @default.
- W3204676188 cites W2963960318 @default.
- W3204676188 cites W2964105864 @default.
- W3204676188 cites W2964159205 @default.
- W3204676188 cites W2968483100 @default.
- W3204676188 cites W2979579363 @default.
- W3204676188 cites W2979689312 @default.
- W3204676188 cites W2982049331 @default.
- W3204676188 cites W2988205463 @default.
- W3204676188 cites W2997006708 @default.
- W3204676188 cites W3009081299 @default.
- W3204676188 cites W3012255272 @default.
- W3204676188 cites W3035524453 @default.
- W3204676188 cites W3171007011 @default.
- W3204676188 cites W3175577495 @default.
- W3204676188 cites W343636949 @default.
- W3204676188 doi "https://doi.org/10.1109/iccv48922.2021.00858" @default.
- W3204676188 hasPublicationYear "2021" @default.
- W3204676188 type Work @default.
- W3204676188 sameAs 3204676188 @default.
- W3204676188 citedByCount "1" @default.
- W3204676188 countsByYear W32046761882023 @default.
- W3204676188 crossrefType "proceedings-article" @default.
- W3204676188 hasAuthorship W3204676188A5033229022 @default.
- W3204676188 hasAuthorship W3204676188A5037727565 @default.
- W3204676188 hasAuthorship W3204676188A5039231803 @default.
- W3204676188 hasAuthorship W3204676188A5062546146 @default.
- W3204676188 hasAuthorship W3204676188A5071983998 @default.
- W3204676188 hasAuthorship W3204676188A5075246991 @default.
- W3204676188 hasConcept C111919701 @default.
- W3204676188 hasConcept C119857082 @default.
- W3204676188 hasConcept C154945302 @default.
- W3204676188 hasConcept C2777212361 @default.
- W3204676188 hasConcept C2911011789 @default.
- W3204676188 hasConcept C41008148 @default.
- W3204676188 hasConcept C98045186 @default.
- W3204676188 hasConceptScore W3204676188C111919701 @default.
- W3204676188 hasConceptScore W3204676188C119857082 @default.
- W3204676188 hasConceptScore W3204676188C154945302 @default.
- W3204676188 hasConceptScore W3204676188C2777212361 @default.
- W3204676188 hasConceptScore W3204676188C2911011789 @default.
- W3204676188 hasConceptScore W3204676188C41008148 @default.
- W3204676188 hasConceptScore W3204676188C98045186 @default.
- W3204676188 hasLocation W32046761881 @default.
- W3204676188 hasOpenAccess W3204676188 @default.
- W3204676188 hasPrimaryLocation W32046761881 @default.
- W3204676188 hasRelatedWork W1498781289 @default.
- W3204676188 hasRelatedWork W1972002222 @default.
- W3204676188 hasRelatedWork W2122243982 @default.
- W3204676188 hasRelatedWork W2134545995 @default.
- W3204676188 hasRelatedWork W2961085424 @default.
- W3204676188 hasRelatedWork W2966288165 @default.
- W3204676188 hasRelatedWork W4285782376 @default.
- W3204676188 hasRelatedWork W4286629047 @default.
- W3204676188 hasRelatedWork W4306674287 @default.
- W3204676188 hasRelatedWork W4224009465 @default.
- W3204676188 isParatext "false" @default.
- W3204676188 isRetracted "false" @default.
- W3204676188 magId "3204676188" @default.
- W3204676188 workType "article" @default.