Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310333713> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W4310333713 endingPage "12" @default.
- W4310333713 startingPage "1" @default.
- W4310333713 abstract "Data augmentation (DA) is a widely used technique for enhancing the training of deep neural networks. Recent DA techniques which achieve state-of-the-art performance always meet the need for diversity in augmented training samples. However, an augmentation strategy that has a high diversity usually introduces out-of-distribution (OOD) augmented samples and these samples consequently impair the performance. To alleviate this issue, we propose ReSmooth, a framework that first detects OOD samples in augmented samples and then leverages them. To be specific, we first use a Gaussian mixture model (GMM) to fit the loss distribution of both the original and augmented samples and accordingly split these samples into in-distribution (ID) samples and OOD samples. Then we start a new training where ID and OOD samples are incorporated with different smooth labels. By treating ID samples and OOD samples unequally, we can make better use of the diverse augmented data. Furthermore, we incorporate our ReSmooth framework with negative DA (NDA) strategies. By properly handling their intentionally created OOD samples, the classification performance of NDAs is largely ameliorated. Experiments on several classification benchmarks show that ReSmooth can be easily extended to the existing augmentation strategies such as RandAugment (RA), rotate, and jigsaw and improve on them. Our code is available at https://github.com/Chenyang4/ReSmooth." @default.
- W4310333713 created "2022-12-08" @default.
- W4310333713 creator A5027059038 @default.
- W4310333713 creator A5040922635 @default.
- W4310333713 creator A5066252438 @default.
- W4310333713 creator A5087165831 @default.
- W4310333713 date "2022-01-01" @default.
- W4310333713 modified "2023-10-16" @default.
- W4310333713 title "ReSmooth: Detecting and Utilizing OOD Samples When Training With Data Augmentation" @default.
- W4310333713 doi "https://doi.org/10.1109/tnnls.2022.3222044" @default.
- W4310333713 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36417730" @default.
- W4310333713 hasPublicationYear "2022" @default.
- W4310333713 type Work @default.
- W4310333713 citedByCount "0" @default.
- W4310333713 crossrefType "journal-article" @default.
- W4310333713 hasAuthorship W4310333713A5027059038 @default.
- W4310333713 hasAuthorship W4310333713A5040922635 @default.
- W4310333713 hasAuthorship W4310333713A5066252438 @default.
- W4310333713 hasAuthorship W4310333713A5087165831 @default.
- W4310333713 hasBestOaLocation W43103337132 @default.
- W4310333713 hasConcept C119857082 @default.
- W4310333713 hasConcept C121332964 @default.
- W4310333713 hasConcept C124101348 @default.
- W4310333713 hasConcept C145420912 @default.
- W4310333713 hasConcept C153180895 @default.
- W4310333713 hasConcept C154945302 @default.
- W4310333713 hasConcept C163716315 @default.
- W4310333713 hasConcept C177264268 @default.
- W4310333713 hasConcept C199360897 @default.
- W4310333713 hasConcept C2776760102 @default.
- W4310333713 hasConcept C2779405079 @default.
- W4310333713 hasConcept C33923547 @default.
- W4310333713 hasConcept C41008148 @default.
- W4310333713 hasConcept C50644808 @default.
- W4310333713 hasConcept C51632099 @default.
- W4310333713 hasConcept C61224824 @default.
- W4310333713 hasConcept C62520636 @default.
- W4310333713 hasConceptScore W4310333713C119857082 @default.
- W4310333713 hasConceptScore W4310333713C121332964 @default.
- W4310333713 hasConceptScore W4310333713C124101348 @default.
- W4310333713 hasConceptScore W4310333713C145420912 @default.
- W4310333713 hasConceptScore W4310333713C153180895 @default.
- W4310333713 hasConceptScore W4310333713C154945302 @default.
- W4310333713 hasConceptScore W4310333713C163716315 @default.
- W4310333713 hasConceptScore W4310333713C177264268 @default.
- W4310333713 hasConceptScore W4310333713C199360897 @default.
- W4310333713 hasConceptScore W4310333713C2776760102 @default.
- W4310333713 hasConceptScore W4310333713C2779405079 @default.
- W4310333713 hasConceptScore W4310333713C33923547 @default.
- W4310333713 hasConceptScore W4310333713C41008148 @default.
- W4310333713 hasConceptScore W4310333713C50644808 @default.
- W4310333713 hasConceptScore W4310333713C51632099 @default.
- W4310333713 hasConceptScore W4310333713C61224824 @default.
- W4310333713 hasConceptScore W4310333713C62520636 @default.
- W4310333713 hasFunder F4320321001 @default.
- W4310333713 hasFunder F4320335777 @default.
- W4310333713 hasFunder F4320335787 @default.
- W4310333713 hasLocation W43103337131 @default.
- W4310333713 hasLocation W43103337132 @default.
- W4310333713 hasLocation W43103337133 @default.
- W4310333713 hasLocation W43103337134 @default.
- W4310333713 hasOpenAccess W4310333713 @default.
- W4310333713 hasPrimaryLocation W43103337131 @default.
- W4310333713 hasRelatedWork W1562793155 @default.
- W4310333713 hasRelatedWork W1597501978 @default.
- W4310333713 hasRelatedWork W2003627431 @default.
- W4310333713 hasRelatedWork W2076520961 @default.
- W4310333713 hasRelatedWork W2107898784 @default.
- W4310333713 hasRelatedWork W2433217581 @default.
- W4310333713 hasRelatedWork W2752848712 @default.
- W4310333713 hasRelatedWork W2954309397 @default.
- W4310333713 hasRelatedWork W2961085424 @default.
- W4310333713 hasRelatedWork W4306674287 @default.
- W4310333713 isParatext "false" @default.
- W4310333713 isRetracted "false" @default.
- W4310333713 workType "article" @default.