FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3119342105> ?p ?o ?g. }

• W3119342105 abstract "As the state-of-the-art machine learning methods in many fields rely on larger datasets, storing them and training models on them becomes more expensive. This paper proposes a training set synthesis technique for emph{data-efficient} learning, called emph{Dataset Condensation}, that learns to condense a large dataset into a small set of informative samples for training deep neural networks from scratch. We formulate this goal as a gradient matching problem between the gradients of a deep neural network trained on the original data and our synthetic data. We rigorously evaluate its performance in several computer vision benchmarks and demonstrate that it significantly outperforms the state-of-the-art methods. Finally we explore the use of our method in continual learning and neural architecture search and show that it achieves promising gains on a tight budget of memory and computations." @default.
• W3119342105 created "2021-01-18" @default.
• W3119342105 creator A5024899038 @default.
• W3119342105 creator A5056977175 @default.
• W3119342105 creator A5064839560 @default.
• W3119342105 date "2021-05-03" @default.
• W3119342105 modified "2023-09-24" @default.
• W3119342105 title "Dataset Condensation with Gradient Matching" @default.
• W3119342105 cites W1665214252 @default.
• W3119342105 cites W1677182931 @default.
• W3119342105 cites W1686810756 @default.
• W3119342105 cites W1821462560 @default.
• W3119342105 cites W1849277567 @default.
• W3119342105 cites W1915485278 @default.
• W3119342105 cites W1959608418 @default.
• W3119342105 cites W1981313592 @default.
• W3119342105 cites W1997865285 @default.
• W3119342105 cites W2035424729 @default.
• W3119342105 cites W2045964207 @default.
• W3119342105 cites W2099471712 @default.
• W3119342105 cites W2100659887 @default.
• W3119342105 cites W2108598243 @default.
• W3119342105 cites W2112796928 @default.
• W3119342105 cites W2125389028 @default.
• W3119342105 cites W2130248603 @default.
• W3119342105 cites W2173520492 @default.
• W3119342105 cites W2187061624 @default.
• W3119342105 cites W2194775991 @default.
• W3119342105 cites W2229238337 @default.
• W3119342105 cites W2335728318 @default.
• W3119342105 cites W2502312327 @default.
• W3119342105 cites W2597603852 @default.
• W3119342105 cites W2601450892 @default.
• W3119342105 cites W2750384547 @default.
• W3119342105 cites W2766966408 @default.
• W3119342105 cites W2798836702 @default.
• W3119342105 cites W2884282566 @default.
• W3119342105 cites W2945528222 @default.
• W3119342105 cites W2949117887 @default.
• W3119342105 cites W2950320139 @default.
• W3119342105 cites W2951013084 @default.
• W3119342105 cites W2954929116 @default.
• W3119342105 cites W2955087950 @default.
• W3119342105 cites W2962724315 @default.
• W3119342105 cites W2963207607 @default.
• W3119342105 cites W2963341924 @default.
• W3119342105 cites W2963902936 @default.
• W3119342105 cites W2964118293 @default.
• W3119342105 cites W2964189064 @default.
• W3119342105 cites W2964515685 @default.
• W3119342105 cites W2970066656 @default.
• W3119342105 cites W2970408908 @default.
• W3119342105 cites W2983647115 @default.
• W3119342105 cites W3000479830 @default.
• W3119342105 cites W3024202154 @default.
• W3119342105 cites W3034828529 @default.
• W3119342105 cites W3035559424 @default.
• W3119342105 cites W3048813204 @default.
• W3119342105 cites W3086152015 @default.
• W3119342105 cites W3118608800 @default.
• W3119342105 cites W3464401 @default.
• W3119342105 hasPublicationYear "2021" @default.
• W3119342105 type Work @default.
• W3119342105 sameAs 3119342105 @default.
• W3119342105 citedByCount "8" @default.
• W3119342105 countsByYear W31193421052020 @default.
• W3119342105 countsByYear W31193421052021 @default.
• W3119342105 crossrefType "proceedings-article" @default.
• W3119342105 hasAuthorship W3119342105A5024899038 @default.
• W3119342105 hasAuthorship W3119342105A5056977175 @default.
• W3119342105 hasAuthorship W3119342105A5064839560 @default.
• W3119342105 hasConcept C105795698 @default.
• W3119342105 hasConcept C108583219 @default.
• W3119342105 hasConcept C111919701 @default.
• W3119342105 hasConcept C11413529 @default.
• W3119342105 hasConcept C119857082 @default.
• W3119342105 hasConcept C123657996 @default.
• W3119342105 hasConcept C142362112 @default.
• W3119342105 hasConcept C153180895 @default.
• W3119342105 hasConcept C153349607 @default.
• W3119342105 hasConcept C154945302 @default.
• W3119342105 hasConcept C165064840 @default.
• W3119342105 hasConcept C177264268 @default.
• W3119342105 hasConcept C199360897 @default.
• W3119342105 hasConcept C2781235140 @default.
• W3119342105 hasConcept C2984842247 @default.
• W3119342105 hasConcept C33923547 @default.
• W3119342105 hasConcept C41008148 @default.
• W3119342105 hasConcept C45374587 @default.
• W3119342105 hasConcept C48103436 @default.
• W3119342105 hasConcept C50644808 @default.
• W3119342105 hasConcept C51632099 @default.
• W3119342105 hasConcept C58489278 @default.
• W3119342105 hasConceptScore W3119342105C105795698 @default.
• W3119342105 hasConceptScore W3119342105C108583219 @default.
• W3119342105 hasConceptScore W3119342105C111919701 @default.
• W3119342105 hasConceptScore W3119342105C11413529 @default.
• W3119342105 hasConceptScore W3119342105C119857082 @default.
• W3119342105 hasConceptScore W3119342105C123657996 @default.
• W3119342105 hasConceptScore W3119342105C142362112 @default.

• next