Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384080802> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4384080802 endingPage "13" @default.
- W4384080802 startingPage "1" @default.
- W4384080802 abstract "Most existing methods that cope with noisy labels usually assume that the classwise data distributions are well balanced. They are difficult to deal with the practical scenarios where training samples have imbalanced distributions, since they are not able to differentiate noisy samples from tail classes' clean samples. This article makes an early effort to tackle the image classification task in which the provided labels are noisy and have a long-tailed distribution. To deal with this problem, we propose a new learning paradigm which can screen out noisy samples by matching between inferences on weak and strong data augmentations. A leave-noise-out regularization (LNOR) is further introduced to eliminate the effect of the recognized noisy samples. Besides, we propose a prediction penalty based on the online classwise confidence levels to avoid the bias toward easy classes which are dominated by head classes. Extensive experiments on five datasets including CIFAR-10, CIFAR-100, MNIST, FashionMNIST, and Clothing1M demonstrate that the proposed method outperforms the existing algorithms for learning with long-tailed distribution and label noise." @default.
- W4384080802 created "2023-07-13" @default.
- W4384080802 creator A5003019748 @default.
- W4384080802 creator A5009740359 @default.
- W4384080802 creator A5043089518 @default.
- W4384080802 creator A5043494334 @default.
- W4384080802 creator A5046465212 @default.
- W4384080802 creator A5050630882 @default.
- W4384080802 creator A5060524512 @default.
- W4384080802 creator A5084779782 @default.
- W4384080802 date "2023-01-01" @default.
- W4384080802 modified "2023-10-15" @default.
- W4384080802 title "Separating Noisy Samples From Tail Classes for Long-Tailed Image Classification With Label Noise" @default.
- W4384080802 doi "https://doi.org/10.1109/tnnls.2023.3291695" @default.
- W4384080802 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37436859" @default.
- W4384080802 hasPublicationYear "2023" @default.
- W4384080802 type Work @default.
- W4384080802 citedByCount "0" @default.
- W4384080802 crossrefType "journal-article" @default.
- W4384080802 hasAuthorship W4384080802A5003019748 @default.
- W4384080802 hasAuthorship W4384080802A5009740359 @default.
- W4384080802 hasAuthorship W4384080802A5043089518 @default.
- W4384080802 hasAuthorship W4384080802A5043494334 @default.
- W4384080802 hasAuthorship W4384080802A5046465212 @default.
- W4384080802 hasAuthorship W4384080802A5050630882 @default.
- W4384080802 hasAuthorship W4384080802A5060524512 @default.
- W4384080802 hasAuthorship W4384080802A5084779782 @default.
- W4384080802 hasConcept C105795698 @default.
- W4384080802 hasConcept C108583219 @default.
- W4384080802 hasConcept C115961682 @default.
- W4384080802 hasConcept C119857082 @default.
- W4384080802 hasConcept C153180895 @default.
- W4384080802 hasConcept C154945302 @default.
- W4384080802 hasConcept C162324750 @default.
- W4384080802 hasConcept C165064840 @default.
- W4384080802 hasConcept C187736073 @default.
- W4384080802 hasConcept C190502265 @default.
- W4384080802 hasConcept C2776135515 @default.
- W4384080802 hasConcept C2780451532 @default.
- W4384080802 hasConcept C2781170535 @default.
- W4384080802 hasConcept C33923547 @default.
- W4384080802 hasConcept C41008148 @default.
- W4384080802 hasConcept C99498987 @default.
- W4384080802 hasConceptScore W4384080802C105795698 @default.
- W4384080802 hasConceptScore W4384080802C108583219 @default.
- W4384080802 hasConceptScore W4384080802C115961682 @default.
- W4384080802 hasConceptScore W4384080802C119857082 @default.
- W4384080802 hasConceptScore W4384080802C153180895 @default.
- W4384080802 hasConceptScore W4384080802C154945302 @default.
- W4384080802 hasConceptScore W4384080802C162324750 @default.
- W4384080802 hasConceptScore W4384080802C165064840 @default.
- W4384080802 hasConceptScore W4384080802C187736073 @default.
- W4384080802 hasConceptScore W4384080802C190502265 @default.
- W4384080802 hasConceptScore W4384080802C2776135515 @default.
- W4384080802 hasConceptScore W4384080802C2780451532 @default.
- W4384080802 hasConceptScore W4384080802C2781170535 @default.
- W4384080802 hasConceptScore W4384080802C33923547 @default.
- W4384080802 hasConceptScore W4384080802C41008148 @default.
- W4384080802 hasConceptScore W4384080802C99498987 @default.
- W4384080802 hasFunder F4320321001 @default.
- W4384080802 hasFunder F4320327912 @default.
- W4384080802 hasLocation W43840808021 @default.
- W4384080802 hasLocation W43840808022 @default.
- W4384080802 hasOpenAccess W4384080802 @default.
- W4384080802 hasPrimaryLocation W43840808021 @default.
- W4384080802 hasRelatedWork W2597787948 @default.
- W4384080802 hasRelatedWork W2810865670 @default.
- W4384080802 hasRelatedWork W2904927891 @default.
- W4384080802 hasRelatedWork W2947175736 @default.
- W4384080802 hasRelatedWork W2951786554 @default.
- W4384080802 hasRelatedWork W2978290780 @default.
- W4384080802 hasRelatedWork W3156786002 @default.
- W4384080802 hasRelatedWork W3177008965 @default.
- W4384080802 hasRelatedWork W3186840088 @default.
- W4384080802 hasRelatedWork W4220663171 @default.
- W4384080802 isParatext "false" @default.
- W4384080802 isRetracted "false" @default.
- W4384080802 workType "article" @default.