Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386644702> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4386644702 abstract "Data augmentation (DA) is widely used to improve the generalization of neural networks by enforcing the invariances and symmetries to pre-defined transformations applied to input data. However, a fixed augmentation policy may have different effects on each sample in different training stages but existing approaches cannot adjust the policy to be adaptive to each sample and the training model. In this paper, we propose Model Adaptive Data Augmentation (MADAug) that jointly trains an augmentation policy network to teach the model when to learn what. Unlike previous work, MADAug selects augmentation operators for each input image by a model-adaptive policy varying between training stages, producing a data augmentation curriculum optimized for better generalization. In MADAug, we train the policy through a bi-level optimization scheme, which aims to minimize a validation-set loss of a model trained using the policy-produced data augmentations. We conduct an extensive evaluation of MADAug on multiple image classification tasks and network architectures with thorough comparisons to existing DA approaches. MADAug outperforms or is on par with other baselines and exhibits better fairness: it brings improvement to all classes and more to the difficult ones. Moreover, MADAug learned policy shows better performance when transferred to fine-grained datasets. In addition, the auto-optimized policy in MADAug gradually introduces increasing perturbations and naturally forms an easy-to-hard curriculum." @default.
- W4386644702 created "2023-09-13" @default.
- W4386644702 creator A5017743340 @default.
- W4386644702 creator A5039076312 @default.
- W4386644702 creator A5073304731 @default.
- W4386644702 date "2023-09-09" @default.
- W4386644702 modified "2023-10-05" @default.
- W4386644702 title "When to Learn What: Model-Adaptive Data Augmentation Curriculum" @default.
- W4386644702 doi "https://doi.org/10.48550/arxiv.2309.04747" @default.
- W4386644702 hasPublicationYear "2023" @default.
- W4386644702 type Work @default.
- W4386644702 citedByCount "0" @default.
- W4386644702 crossrefType "posted-content" @default.
- W4386644702 hasAuthorship W4386644702A5017743340 @default.
- W4386644702 hasAuthorship W4386644702A5039076312 @default.
- W4386644702 hasAuthorship W4386644702A5073304731 @default.
- W4386644702 hasBestOaLocation W43866447021 @default.
- W4386644702 hasConcept C119857082 @default.
- W4386644702 hasConcept C134306372 @default.
- W4386644702 hasConcept C154945302 @default.
- W4386644702 hasConcept C15744967 @default.
- W4386644702 hasConcept C177148314 @default.
- W4386644702 hasConcept C177264268 @default.
- W4386644702 hasConcept C185592680 @default.
- W4386644702 hasConcept C190839683 @default.
- W4386644702 hasConcept C19417346 @default.
- W4386644702 hasConcept C198531522 @default.
- W4386644702 hasConcept C199360897 @default.
- W4386644702 hasConcept C205649164 @default.
- W4386644702 hasConcept C33923547 @default.
- W4386644702 hasConcept C41008148 @default.
- W4386644702 hasConcept C43617362 @default.
- W4386644702 hasConcept C47177190 @default.
- W4386644702 hasConcept C50644808 @default.
- W4386644702 hasConcept C58640448 @default.
- W4386644702 hasConcept C77618280 @default.
- W4386644702 hasConceptScore W4386644702C119857082 @default.
- W4386644702 hasConceptScore W4386644702C134306372 @default.
- W4386644702 hasConceptScore W4386644702C154945302 @default.
- W4386644702 hasConceptScore W4386644702C15744967 @default.
- W4386644702 hasConceptScore W4386644702C177148314 @default.
- W4386644702 hasConceptScore W4386644702C177264268 @default.
- W4386644702 hasConceptScore W4386644702C185592680 @default.
- W4386644702 hasConceptScore W4386644702C190839683 @default.
- W4386644702 hasConceptScore W4386644702C19417346 @default.
- W4386644702 hasConceptScore W4386644702C198531522 @default.
- W4386644702 hasConceptScore W4386644702C199360897 @default.
- W4386644702 hasConceptScore W4386644702C205649164 @default.
- W4386644702 hasConceptScore W4386644702C33923547 @default.
- W4386644702 hasConceptScore W4386644702C41008148 @default.
- W4386644702 hasConceptScore W4386644702C43617362 @default.
- W4386644702 hasConceptScore W4386644702C47177190 @default.
- W4386644702 hasConceptScore W4386644702C50644808 @default.
- W4386644702 hasConceptScore W4386644702C58640448 @default.
- W4386644702 hasConceptScore W4386644702C77618280 @default.
- W4386644702 hasLocation W43866447021 @default.
- W4386644702 hasOpenAccess W4386644702 @default.
- W4386644702 hasPrimaryLocation W43866447021 @default.
- W4386644702 hasRelatedWork W1568097102 @default.
- W4386644702 hasRelatedWork W1601203902 @default.
- W4386644702 hasRelatedWork W2030078074 @default.
- W4386644702 hasRelatedWork W2248934910 @default.
- W4386644702 hasRelatedWork W2361332776 @default.
- W4386644702 hasRelatedWork W2377336366 @default.
- W4386644702 hasRelatedWork W2897407000 @default.
- W4386644702 hasRelatedWork W2899084033 @default.
- W4386644702 hasRelatedWork W4225671779 @default.
- W4386644702 hasRelatedWork W618248309 @default.
- W4386644702 isParatext "false" @default.
- W4386644702 isRetracted "false" @default.
- W4386644702 workType "article" @default.