Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387323198> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4387323198 abstract "Researchers have long tried to minimize training costs in deep learning while maintaining strong generalization across diverse datasets. Emerging research on dataset distillation aims to reduce training costs by creating a small synthetic set that contains the information of a larger real dataset and ultimately achieves test accuracy equivalent to a model trained on the whole dataset. Unfortunately, the synthetic data generated by previous methods are not guaranteed to distribute and discriminate as well as the original training data, and they incur significant computational costs. Despite promising results, there still exists a significant performance gap between models trained on condensed synthetic sets and those trained on the whole dataset. In this paper, we address these challenges using efficient Dataset Distillation with Attention Matching (DataDAM), achieving state-of-the-art performance while reducing training costs. Specifically, we learn synthetic images by matching the spatial attention maps of real and synthetic data generated by different layers within a family of randomly initialized neural networks. Our method outperforms the prior methods on several datasets, including CIFAR10/100, TinyImageNet, ImageNet-1K, and subsets of ImageNet-1K across most of the settings, and achieves improvements of up to 6.5% and 4.1% on CIFAR100 and ImageNet-1K, respectively. We also show that our high-quality distilled images have practical benefits for downstream applications, such as continual learning and neural architecture search." @default.
- W4387323198 created "2023-10-04" @default.
- W4387323198 creator A5036842123 @default.
- W4387323198 creator A5059152392 @default.
- W4387323198 creator A5067534714 @default.
- W4387323198 creator A5074794416 @default.
- W4387323198 creator A5079906820 @default.
- W4387323198 creator A5082010781 @default.
- W4387323198 date "2023-09-29" @default.
- W4387323198 modified "2023-10-05" @default.
- W4387323198 title "DataDAM: Efficient Dataset Distillation with Attention Matching" @default.
- W4387323198 doi "https://doi.org/10.48550/arxiv.2310.00093" @default.
- W4387323198 hasPublicationYear "2023" @default.
- W4387323198 type Work @default.
- W4387323198 citedByCount "0" @default.
- W4387323198 crossrefType "posted-content" @default.
- W4387323198 hasAuthorship W4387323198A5036842123 @default.
- W4387323198 hasAuthorship W4387323198A5059152392 @default.
- W4387323198 hasAuthorship W4387323198A5067534714 @default.
- W4387323198 hasAuthorship W4387323198A5074794416 @default.
- W4387323198 hasAuthorship W4387323198A5079906820 @default.
- W4387323198 hasAuthorship W4387323198A5082010781 @default.
- W4387323198 hasBestOaLocation W43873231981 @default.
- W4387323198 hasConcept C105795698 @default.
- W4387323198 hasConcept C119857082 @default.
- W4387323198 hasConcept C124101348 @default.
- W4387323198 hasConcept C134306372 @default.
- W4387323198 hasConcept C153180895 @default.
- W4387323198 hasConcept C154945302 @default.
- W4387323198 hasConcept C160920958 @default.
- W4387323198 hasConcept C165064840 @default.
- W4387323198 hasConcept C169903167 @default.
- W4387323198 hasConcept C177148314 @default.
- W4387323198 hasConcept C177264268 @default.
- W4387323198 hasConcept C178790620 @default.
- W4387323198 hasConcept C185592680 @default.
- W4387323198 hasConcept C199360897 @default.
- W4387323198 hasConcept C204030448 @default.
- W4387323198 hasConcept C33923547 @default.
- W4387323198 hasConcept C41008148 @default.
- W4387323198 hasConcept C50644808 @default.
- W4387323198 hasConcept C51632099 @default.
- W4387323198 hasConceptScore W4387323198C105795698 @default.
- W4387323198 hasConceptScore W4387323198C119857082 @default.
- W4387323198 hasConceptScore W4387323198C124101348 @default.
- W4387323198 hasConceptScore W4387323198C134306372 @default.
- W4387323198 hasConceptScore W4387323198C153180895 @default.
- W4387323198 hasConceptScore W4387323198C154945302 @default.
- W4387323198 hasConceptScore W4387323198C160920958 @default.
- W4387323198 hasConceptScore W4387323198C165064840 @default.
- W4387323198 hasConceptScore W4387323198C169903167 @default.
- W4387323198 hasConceptScore W4387323198C177148314 @default.
- W4387323198 hasConceptScore W4387323198C177264268 @default.
- W4387323198 hasConceptScore W4387323198C178790620 @default.
- W4387323198 hasConceptScore W4387323198C185592680 @default.
- W4387323198 hasConceptScore W4387323198C199360897 @default.
- W4387323198 hasConceptScore W4387323198C204030448 @default.
- W4387323198 hasConceptScore W4387323198C33923547 @default.
- W4387323198 hasConceptScore W4387323198C41008148 @default.
- W4387323198 hasConceptScore W4387323198C50644808 @default.
- W4387323198 hasConceptScore W4387323198C51632099 @default.
- W4387323198 hasLocation W43873231981 @default.
- W4387323198 hasOpenAccess W4387323198 @default.
- W4387323198 hasPrimaryLocation W43873231981 @default.
- W4387323198 hasRelatedWork W17944005 @default.
- W4387323198 hasRelatedWork W1972167985 @default.
- W4387323198 hasRelatedWork W1999699871 @default.
- W4387323198 hasRelatedWork W2006801911 @default.
- W4387323198 hasRelatedWork W2021633306 @default.
- W4387323198 hasRelatedWork W2043806667 @default.
- W4387323198 hasRelatedWork W2971899271 @default.
- W4387323198 hasRelatedWork W3175965105 @default.
- W4387323198 hasRelatedWork W4225124612 @default.
- W4387323198 hasRelatedWork W4312659495 @default.
- W4387323198 isParatext "false" @default.
- W4387323198 isRetracted "false" @default.
- W4387323198 workType "article" @default.