SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3128017834> ?p ?o ?g. }

Showing items 1 to 79 of 79 with 100 items per page.

W3128017834 abstract "Few-shot learning aims to train a classifier given only a few samples per class that are highly insufficient to describe the whole data distribution. These few-shot samples not only introduce high variance to the training but also may include outliers near the class boundaries. Directly feeding these samples to training algorithms can lead to unstable optimization and even incorrect gradient descent direction. In this paper, we improve the robustness to ``outliers'' by learning to propagate and refine the representations of few-shot samples to form a more compact data distribution before using them to train a classifier. We develop a mutual calibration among few-shot samples' representations by graph propagation, for which we learn an attention mechanism to build the graph and determine the propagation weights. On both clean datasets and datasets containing noisy labels, we show that our sample propagation generally improves different types of existing few-shot learning methods in multiple few-shot learning settings." @default.
W3128017834 created "2021-02-15" @default.
W3128017834 creator A5039076312 @default.
W3128017834 creator A5049264813 @default.
W3128017834 creator A5057139422 @default.
W3128017834 creator A5059227406 @default.
W3128017834 creator A5069837273 @default.
W3128017834 creator A5074852078 @default.
W3128017834 date "2021-05-04" @default.
W3128017834 modified "2023-09-23" @default.
W3128017834 title "MASP: Model-Agnostic Sample Propagation for Few-shot learning" @default.
W3128017834 hasPublicationYear "2021" @default.
W3128017834 type Work @default.
W3128017834 sameAs 3128017834 @default.
W3128017834 citedByCount "0" @default.
W3128017834 crossrefType "journal-article" @default.
W3128017834 hasAuthorship W3128017834A5039076312 @default.
W3128017834 hasAuthorship W3128017834A5049264813 @default.
W3128017834 hasAuthorship W3128017834A5057139422 @default.
W3128017834 hasAuthorship W3128017834A5059227406 @default.
W3128017834 hasAuthorship W3128017834A5069837273 @default.
W3128017834 hasAuthorship W3128017834A5074852078 @default.
W3128017834 hasConcept C104317684 @default.
W3128017834 hasConcept C119857082 @default.
W3128017834 hasConcept C124101348 @default.
W3128017834 hasConcept C132525143 @default.
W3128017834 hasConcept C153180895 @default.
W3128017834 hasConcept C153258448 @default.
W3128017834 hasConcept C154945302 @default.
W3128017834 hasConcept C185592680 @default.
W3128017834 hasConcept C41008148 @default.
W3128017834 hasConcept C50644808 @default.
W3128017834 hasConcept C55493867 @default.
W3128017834 hasConcept C63479239 @default.
W3128017834 hasConcept C79337645 @default.
W3128017834 hasConcept C80444323 @default.
W3128017834 hasConcept C95623464 @default.
W3128017834 hasConceptScore W3128017834C104317684 @default.
W3128017834 hasConceptScore W3128017834C119857082 @default.
W3128017834 hasConceptScore W3128017834C124101348 @default.
W3128017834 hasConceptScore W3128017834C132525143 @default.
W3128017834 hasConceptScore W3128017834C153180895 @default.
W3128017834 hasConceptScore W3128017834C153258448 @default.
W3128017834 hasConceptScore W3128017834C154945302 @default.
W3128017834 hasConceptScore W3128017834C185592680 @default.
W3128017834 hasConceptScore W3128017834C41008148 @default.
W3128017834 hasConceptScore W3128017834C50644808 @default.
W3128017834 hasConceptScore W3128017834C55493867 @default.
W3128017834 hasConceptScore W3128017834C63479239 @default.
W3128017834 hasConceptScore W3128017834C79337645 @default.
W3128017834 hasConceptScore W3128017834C80444323 @default.
W3128017834 hasConceptScore W3128017834C95623464 @default.
W3128017834 hasLocation W31280178341 @default.
W3128017834 hasOpenAccess W3128017834 @default.
W3128017834 hasPrimaryLocation W31280178341 @default.
W3128017834 hasRelatedWork W1943830695 @default.
W3128017834 hasRelatedWork W2107483511 @default.
W3128017834 hasRelatedWork W229388967 @default.
W3128017834 hasRelatedWork W2315658283 @default.
W3128017834 hasRelatedWork W2889809710 @default.
W3128017834 hasRelatedWork W2897510511 @default.
W3128017834 hasRelatedWork W2898002919 @default.
W3128017834 hasRelatedWork W2929862222 @default.
W3128017834 hasRelatedWork W2972038395 @default.
W3128017834 hasRelatedWork W3036830464 @default.
W3128017834 hasRelatedWork W3101158182 @default.
W3128017834 hasRelatedWork W3137741989 @default.
W3128017834 hasRelatedWork W3161236001 @default.
W3128017834 hasRelatedWork W3162192368 @default.
W3128017834 hasRelatedWork W3174505228 @default.
W3128017834 hasRelatedWork W3182165708 @default.
W3128017834 hasRelatedWork W3189098873 @default.
W3128017834 hasRelatedWork W3194914530 @default.
W3128017834 hasRelatedWork W8597964 @default.
W3128017834 hasRelatedWork W3119279940 @default.
W3128017834 isParatext "false" @default.
W3128017834 isRetracted "false" @default.
W3128017834 magId "3128017834" @default.
W3128017834 workType "article" @default.