FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2910327636> ?p ?o ?g. }

• W2910327636 abstract "Neural architecture search (NAS) methods have been proposed to release human experts from tedious architecture engineering. However, most current methods are constrained in small-scale search due to the issue of computational resources. Meanwhile, directly applying architectures searched on small datasets to large datasets often bears no performance guarantee. This limitation impedes the wide use of NAS on large-scale tasks. To overcome this obstacle, we propose an elastic architecture transfer mechanism for accelerating large-scale neural architecture search (EAT-NAS). In our implementations, architectures are first searched on a small dataset, e.g., CIFAR-10. The best one is chosen as the basic architecture. The search process on the large dataset, e.g., ImageNet, is initialized with the basic architecture as the seed. The large-scale search process is accelerated with the help of the basic architecture. What we propose is not only a NAS method but a mechanism for architecture-level transfer. In our experiments, we obtain two final models EATNet-A and EATNet-B that achieve competitive accuracies, 74.7% and 74.2% on ImageNet, respectively, which also surpass the models searched from scratch on ImageNet under the same settings. For the computational cost, EAT-NAS takes only less than 5 days on 8 TITAN X GPUs, which is significantly less than the computational consumption of the state-of-the-art large-scale NAS methods." @default.
• W2910327636 created "2019-01-25" @default.
• W2910327636 creator A5032279676 @default.
• W2910327636 creator A5037191476 @default.
• W2910327636 creator A5037277702 @default.
• W2910327636 creator A5044759284 @default.
• W2910327636 creator A5052441498 @default.
• W2910327636 creator A5060855982 @default.
• W2910327636 creator A5063536578 @default.
• W2910327636 creator A5087351521 @default.
• W2910327636 date "2019-01-17" @default.
• W2910327636 modified "2023-09-24" @default.
• W2910327636 title "EAT-NAS: Elastic Architecture Transfer for Accelerating Large-scale Neural Architecture Search" @default.
• W2910327636 cites W1568834902 @default.
• W2910327636 cites W2097117768 @default.
• W2910327636 cites W2108598243 @default.
• W2910327636 cites W2163605009 @default.
• W2910327636 cites W2194775991 @default.
• W2910327636 cites W2198098822 @default.
• W2910327636 cites W2531409750 @default.
• W2910327636 cites W2593744649 @default.
• W2910327636 cites W2612445135 @default.
• W2910327636 cites W2746314669 @default.
• W2910327636 cites W2757251151 @default.
• W2910327636 cites W2796265726 @default.
• W2910327636 cites W2810075754 @default.
• W2910327636 cites W2892366900 @default.
• W2910327636 cites W2894957136 @default.
• W2910327636 cites W2962746461 @default.
• W2910327636 cites W2962835968 @default.
• W2910327636 cites W2962847160 @default.
• W2910327636 cites W2963125010 @default.
• W2910327636 cites W2963374479 @default.
• W2910327636 cites W2963628712 @default.
• W2910327636 cites W2963778169 @default.
• W2910327636 cites W2963821229 @default.
• W2910327636 cites W2963918968 @default.
• W2910327636 cites W2964081807 @default.
• W2910327636 cites W2964259004 @default.
• W2910327636 cites W2965658867 @default.
• W2910327636 cites W3118608800 @default.
• W2910327636 hasPublicationYear "2019" @default.
• W2910327636 type Work @default.
• W2910327636 sameAs 2910327636 @default.
• W2910327636 citedByCount "6" @default.
• W2910327636 countsByYear W29103276362019 @default.
• W2910327636 countsByYear W29103276362020 @default.
• W2910327636 crossrefType "posted-content" @default.
• W2910327636 hasAuthorship W2910327636A5032279676 @default.
• W2910327636 hasAuthorship W2910327636A5037191476 @default.
• W2910327636 hasAuthorship W2910327636A5037277702 @default.
• W2910327636 hasAuthorship W2910327636A5044759284 @default.
• W2910327636 hasAuthorship W2910327636A5052441498 @default.
• W2910327636 hasAuthorship W2910327636A5060855982 @default.
• W2910327636 hasAuthorship W2910327636A5063536578 @default.
• W2910327636 hasAuthorship W2910327636A5087351521 @default.
• W2910327636 hasConcept C115903868 @default.
• W2910327636 hasConcept C118524514 @default.
• W2910327636 hasConcept C119857082 @default.
• W2910327636 hasConcept C120314980 @default.
• W2910327636 hasConcept C121332964 @default.
• W2910327636 hasConcept C123657996 @default.
• W2910327636 hasConcept C142362112 @default.
• W2910327636 hasConcept C153349607 @default.
• W2910327636 hasConcept C154945302 @default.
• W2910327636 hasConcept C17744445 @default.
• W2910327636 hasConcept C199360897 @default.
• W2910327636 hasConcept C199539241 @default.
• W2910327636 hasConcept C26713055 @default.
• W2910327636 hasConcept C2776650193 @default.
• W2910327636 hasConcept C2778755073 @default.
• W2910327636 hasConcept C41008148 @default.
• W2910327636 hasConcept C62520636 @default.
• W2910327636 hasConcept C98045186 @default.
• W2910327636 hasConceptScore W2910327636C115903868 @default.
• W2910327636 hasConceptScore W2910327636C118524514 @default.
• W2910327636 hasConceptScore W2910327636C119857082 @default.
• W2910327636 hasConceptScore W2910327636C120314980 @default.
• W2910327636 hasConceptScore W2910327636C121332964 @default.
• W2910327636 hasConceptScore W2910327636C123657996 @default.
• W2910327636 hasConceptScore W2910327636C142362112 @default.
• W2910327636 hasConceptScore W2910327636C153349607 @default.
• W2910327636 hasConceptScore W2910327636C154945302 @default.
• W2910327636 hasConceptScore W2910327636C17744445 @default.
• W2910327636 hasConceptScore W2910327636C199360897 @default.
• W2910327636 hasConceptScore W2910327636C199539241 @default.
• W2910327636 hasConceptScore W2910327636C26713055 @default.
• W2910327636 hasConceptScore W2910327636C2776650193 @default.
• W2910327636 hasConceptScore W2910327636C2778755073 @default.
• W2910327636 hasConceptScore W2910327636C41008148 @default.
• W2910327636 hasConceptScore W2910327636C62520636 @default.
• W2910327636 hasConceptScore W2910327636C98045186 @default.
• W2910327636 hasOpenAccess W2910327636 @default.
• W2910327636 hasRelatedWork W1861492603 @default.
• W2910327636 hasRelatedWork W2108598243 @default.
• W2910327636 hasRelatedWork W2163605009 @default.
• W2910327636 hasRelatedWork W2194775991 @default.
• W2910327636 hasRelatedWork W2553303224 @default.
• W2910327636 hasRelatedWork W2612445135 @default.
• W2910327636 hasRelatedWork W2810075754 @default.

• next