FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313129750> ?p ?o ?g. }

• W4313129750 endingPage "72" @default.
• W4313129750 startingPage "56" @default.
• W4313129750 abstract "Active learning (AL) is a label-efficient technique for training deep models when only a limited labeled set is available and the manual annotation is expensive. Implicit semantic data augmentation (ISDA) effectively extends the limited amount of labeled samples and increases the diversity of labeled sets without introducing a noticeable extra computational cost. The scarcity of labeled instances and the huge annotation cost of unlabelled samples encourage us to ponder on the combination of AL and ISDA. A nature direction is a pipelined integration, which selects the unlabeled samples via acquisition function in AL for labeling and generates virtual samples by changing the selected samples to semantic transformation directions within ISDA. However, this pipelined combination would not guarantee the diversity of virtual samples. This paper proposes diversity-aware semantic transformation active learning, or DAST-AL framework, that looks ahead the effect of ISDA in the selection of unlabeled samples. Specifically, DAST-AL exploits expected partial model change maximization (EPMCM) to consider selected samples’ potential contribution of the diversity to the labeled set by leveraging the semantic transformation within ISDA when selecting the unlabeled samples. After that, DAST-AL can confidently and efficiently augment the labeled set by implicitly generating more diverse samples. The empirical results on both image classification and semantic segmentation tasks show that the proposed DAST-AL can slightly outperform the state-of-the-art AL approaches. Under the same condition, the proposed method takes less than 3 min for the first cycle of active labeling while the existing agreement discrepancy selection incurs more than 40 min." @default.
• W4313129750 created "2023-01-06" @default.
• W4313129750 creator A5008344915 @default.
• W4313129750 creator A5011060162 @default.
• W4313129750 creator A5028146772 @default.
• W4313129750 creator A5047093288 @default.
• W4313129750 creator A5047767249 @default.
• W4313129750 date "2022-01-01" @default.
• W4313129750 modified "2023-10-16" @default.
• W4313129750 title "When Active Learning Meets Implicit Semantic Data Augmentation" @default.
• W4313129750 cites W1955857676 @default.
• W4313129750 cites W1986730033 @default.
• W4313129750 cites W1993615002 @default.
• W4313129750 cites W2042932437 @default.
• W4313129750 cites W2108598243 @default.
• W4313129750 cites W2124244761 @default.
• W4313129750 cites W2194775991 @default.
• W4313129750 cites W2340897893 @default.
• W4313129750 cites W2341531400 @default.
• W4313129750 cites W2740962769 @default.
• W4313129750 cites W2750549109 @default.
• W4313129750 cites W2798820905 @default.
• W4313129750 cites W2883725317 @default.
• W4313129750 cites W2890546378 @default.
• W4313129750 cites W2948729509 @default.
• W4313129750 cites W2951911250 @default.
• W4313129750 cites W2956371155 @default.
• W4313129750 cites W2957775769 @default.
• W4313129750 cites W2962850830 @default.
• W4313129750 cites W2963895143 @default.
• W4313129750 cites W2981952041 @default.
• W4313129750 cites W2986514296 @default.
• W4313129750 cites W2998388430 @default.
• W4313129750 cites W3003262368 @default.
• W4313129750 cites W3025091470 @default.
• W4313129750 cites W3035406940 @default.
• W4313129750 cites W3035499919 @default.
• W4313129750 cites W3166229844 @default.
• W4313129750 cites W3177022620 @default.
• W4313129750 cites W3179550234 @default.
• W4313129750 cites W3184557372 @default.
• W4313129750 cites W343636949 @default.
• W4313129750 doi "https://doi.org/10.1007/978-3-031-19806-9_4" @default.
• W4313129750 hasPublicationYear "2022" @default.
• W4313129750 type Work @default.
• W4313129750 citedByCount "0" @default.
• W4313129750 crossrefType "book-chapter" @default.
• W4313129750 hasAuthorship W4313129750A5008344915 @default.
• W4313129750 hasAuthorship W4313129750A5011060162 @default.
• W4313129750 hasAuthorship W4313129750A5028146772 @default.
• W4313129750 hasAuthorship W4313129750A5047093288 @default.
• W4313129750 hasAuthorship W4313129750A5047767249 @default.
• W4313129750 hasConcept C104317684 @default.
• W4313129750 hasConcept C119857082 @default.
• W4313129750 hasConcept C126255220 @default.
• W4313129750 hasConcept C14036430 @default.
• W4313129750 hasConcept C153180895 @default.
• W4313129750 hasConcept C154945302 @default.
• W4313129750 hasConcept C177264268 @default.
• W4313129750 hasConcept C185592680 @default.
• W4313129750 hasConcept C199360897 @default.
• W4313129750 hasConcept C204241405 @default.
• W4313129750 hasConcept C204321447 @default.
• W4313129750 hasConcept C2776321320 @default.
• W4313129750 hasConcept C2776330181 @default.
• W4313129750 hasConcept C33923547 @default.
• W4313129750 hasConcept C41008148 @default.
• W4313129750 hasConcept C55493867 @default.
• W4313129750 hasConcept C78458016 @default.
• W4313129750 hasConcept C86803240 @default.
• W4313129750 hasConcept C89600930 @default.
• W4313129750 hasConceptScore W4313129750C104317684 @default.
• W4313129750 hasConceptScore W4313129750C119857082 @default.
• W4313129750 hasConceptScore W4313129750C126255220 @default.
• W4313129750 hasConceptScore W4313129750C14036430 @default.
• W4313129750 hasConceptScore W4313129750C153180895 @default.
• W4313129750 hasConceptScore W4313129750C154945302 @default.
• W4313129750 hasConceptScore W4313129750C177264268 @default.
• W4313129750 hasConceptScore W4313129750C185592680 @default.
• W4313129750 hasConceptScore W4313129750C199360897 @default.
• W4313129750 hasConceptScore W4313129750C204241405 @default.
• W4313129750 hasConceptScore W4313129750C204321447 @default.
• W4313129750 hasConceptScore W4313129750C2776321320 @default.
• W4313129750 hasConceptScore W4313129750C2776330181 @default.
• W4313129750 hasConceptScore W4313129750C33923547 @default.
• W4313129750 hasConceptScore W4313129750C41008148 @default.
• W4313129750 hasConceptScore W4313129750C55493867 @default.
• W4313129750 hasConceptScore W4313129750C78458016 @default.
• W4313129750 hasConceptScore W4313129750C86803240 @default.
• W4313129750 hasConceptScore W4313129750C89600930 @default.
• W4313129750 hasLocation W43131297501 @default.
• W4313129750 hasOpenAccess W4313129750 @default.
• W4313129750 hasPrimaryLocation W43131297501 @default.
• W4313129750 hasRelatedWork W1529400504 @default.
• W4313129750 hasRelatedWork W1892467659 @default.
• W4313129750 hasRelatedWork W2961085424 @default.
• W4313129750 hasRelatedWork W3046775127 @default.
• W4313129750 hasRelatedWork W3170094116 @default.

• next