Matches in SemOpenAlex for { <https://semopenalex.org/work/W3127582016> ?p ?o ?g. }
- W3127582016 abstract "Few-shot classification (FSC), the task of adapting a classifier to unseen classes given a small labeled dataset, is an important step on the path toward human-like machine learning. Bayesian methods are well-suited to tackling the fundamental issue of overfitting in the few-shot scenario because they allow practitioners to specify prior beliefs and update those beliefs in light of observed data. Contemporary approaches to Bayesian few-shot classification maintain a posterior distribution over model parameters, which is slow and requires storage that scales with model size. Instead, we propose a Gaussian process classifier based on a novel combination of Polya-gamma augmentation and the one-vs-each softmax approximation that allows us to efficiently marginalize over functions rather than model parameters. We demonstrate improved accuracy and uncertainty quantification on both standard few-shot classification benchmarks and few-shot domain transfer tasks." @default.
- W3127582016 created "2021-02-15" @default.
- W3127582016 creator A5000111344 @default.
- W3127582016 creator A5061896635 @default.
- W3127582016 date "2021-05-03" @default.
- W3127582016 modified "2023-09-23" @default.
- W3127582016 title "Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes" @default.
- W3127582016 cites W137285897 @default.
- W3127582016 cites W1515272691 @default.
- W3127582016 cites W1544760581 @default.
- W3127582016 cites W1605306501 @default.
- W3127582016 cites W1746819321 @default.
- W3127582016 cites W1826234144 @default.
- W3127582016 cites W2001619934 @default.
- W3127582016 cites W2024931169 @default.
- W3127582016 cites W2073241381 @default.
- W3127582016 cites W2099878672 @default.
- W3127582016 cites W2108306139 @default.
- W3127582016 cites W2128973832 @default.
- W3127582016 cites W2133864802 @default.
- W3127582016 cites W2148319469 @default.
- W3127582016 cites W2164411961 @default.
- W3127582016 cites W2401823607 @default.
- W3127582016 cites W2526800167 @default.
- W3127582016 cites W2601450892 @default.
- W3127582016 cites W2604763608 @default.
- W3127582016 cites W2734358244 @default.
- W3127582016 cites W2753160622 @default.
- W3127582016 cites W2786398923 @default.
- W3127582016 cites W2908369924 @default.
- W3127582016 cites W2909021043 @default.
- W3127582016 cites W2947341830 @default.
- W3127582016 cites W2961540362 @default.
- W3127582016 cites W2963341924 @default.
- W3127582016 cites W2963741406 @default.
- W3127582016 cites W2963956018 @default.
- W3127582016 cites W2964078140 @default.
- W3127582016 cites W2964105864 @default.
- W3127582016 cites W2964112702 @default.
- W3127582016 cites W2964121744 @default.
- W3127582016 cites W2964121937 @default.
- W3127582016 cites W2964135722 @default.
- W3127582016 cites W2964212410 @default.
- W3127582016 cites W2964387878 @default.
- W3127582016 cites W2970859221 @default.
- W3127582016 cites W2980037473 @default.
- W3127582016 cites W2989893062 @default.
- W3127582016 cites W3083277700 @default.
- W3127582016 cites W3091905774 @default.
- W3127582016 cites W68072195 @default.
- W3127582016 hasPublicationYear "2021" @default.
- W3127582016 type Work @default.
- W3127582016 sameAs 3127582016 @default.
- W3127582016 citedByCount "4" @default.
- W3127582016 countsByYear W31275820162020 @default.
- W3127582016 countsByYear W31275820162021 @default.
- W3127582016 crossrefType "proceedings-article" @default.
- W3127582016 hasAuthorship W3127582016A5000111344 @default.
- W3127582016 hasAuthorship W3127582016A5061896635 @default.
- W3127582016 hasConcept C107673813 @default.
- W3127582016 hasConcept C108583219 @default.
- W3127582016 hasConcept C119857082 @default.
- W3127582016 hasConcept C121332964 @default.
- W3127582016 hasConcept C127413603 @default.
- W3127582016 hasConcept C153180895 @default.
- W3127582016 hasConcept C154945302 @default.
- W3127582016 hasConcept C163716315 @default.
- W3127582016 hasConcept C178790620 @default.
- W3127582016 hasConcept C185592680 @default.
- W3127582016 hasConcept C188441871 @default.
- W3127582016 hasConcept C22019652 @default.
- W3127582016 hasConcept C2778344882 @default.
- W3127582016 hasConcept C2992734406 @default.
- W3127582016 hasConcept C41008148 @default.
- W3127582016 hasConcept C50644808 @default.
- W3127582016 hasConcept C57830394 @default.
- W3127582016 hasConcept C61326573 @default.
- W3127582016 hasConcept C62520636 @default.
- W3127582016 hasConcept C78519656 @default.
- W3127582016 hasConcept C95623464 @default.
- W3127582016 hasConceptScore W3127582016C107673813 @default.
- W3127582016 hasConceptScore W3127582016C108583219 @default.
- W3127582016 hasConceptScore W3127582016C119857082 @default.
- W3127582016 hasConceptScore W3127582016C121332964 @default.
- W3127582016 hasConceptScore W3127582016C127413603 @default.
- W3127582016 hasConceptScore W3127582016C153180895 @default.
- W3127582016 hasConceptScore W3127582016C154945302 @default.
- W3127582016 hasConceptScore W3127582016C163716315 @default.
- W3127582016 hasConceptScore W3127582016C178790620 @default.
- W3127582016 hasConceptScore W3127582016C185592680 @default.
- W3127582016 hasConceptScore W3127582016C188441871 @default.
- W3127582016 hasConceptScore W3127582016C22019652 @default.
- W3127582016 hasConceptScore W3127582016C2778344882 @default.
- W3127582016 hasConceptScore W3127582016C2992734406 @default.
- W3127582016 hasConceptScore W3127582016C41008148 @default.
- W3127582016 hasConceptScore W3127582016C50644808 @default.
- W3127582016 hasConceptScore W3127582016C57830394 @default.
- W3127582016 hasConceptScore W3127582016C61326573 @default.
- W3127582016 hasConceptScore W3127582016C62520636 @default.
- W3127582016 hasConceptScore W3127582016C78519656 @default.