Matches in SemOpenAlex for { <https://semopenalex.org/work/W3123621629> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3123621629 abstract "Despite the empirical success of neural architecture search (NAS) in deep learning applications, the optimality, reproducibility and cost of NAS schemes remain hard to assess. The variation in search spaces adopted has further affected a fair comparison between search strategies. In this paper, we focus on search strategies in NAS and propose Generative Adversarial NAS (GA-NAS), promoting stable and reproducible neural architecture search. GA-NAS is theoretically inspired by importance sampling for rare event simulation, and iteratively refits a generator to previously discovered top architectures, thus increasingly focusing on important parts of the search space. We propose an efficient adversarial learning approach in GA-NAS, where the generator is not trained based on a large number of observations on architecture performance, but based on the relative prediction made by a discriminator, thus significantly reducing the number of evaluations required. Extensive experiments show that GA-NAS beats the best published results under several cases on the public NAS benchmarks including NAS-Bench-101, NAS-Bench-201, and NAS-Bench-301. We further show that GA-NAS can handle ad-hoc search constraints and search spaces. GA-NAS can find new architectures that enhance EfficientNet and ProxylessNAS in terms of ImageNet Top-1 accuracy and/or the number of parameters by searching in their original search spaces." @default.
- W3123621629 created "2021-02-01" @default.
- W3123621629 creator A5006778208 @default.
- W3123621629 creator A5012650575 @default.
- W3123621629 creator A5032424832 @default.
- W3123621629 creator A5044090050 @default.
- W3123621629 creator A5051959827 @default.
- W3123621629 creator A5083314813 @default.
- W3123621629 date "2021-05-04" @default.
- W3123621629 modified "2023-09-27" @default.
- W3123621629 title "Generative Adversarial Neural Architecture Search with Importance Sampling" @default.
- W3123621629 hasPublicationYear "2021" @default.
- W3123621629 type Work @default.
- W3123621629 sameAs 3123621629 @default.
- W3123621629 citedByCount "0" @default.
- W3123621629 crossrefType "journal-article" @default.
- W3123621629 hasAuthorship W3123621629A5006778208 @default.
- W3123621629 hasAuthorship W3123621629A5012650575 @default.
- W3123621629 hasAuthorship W3123621629A5032424832 @default.
- W3123621629 hasAuthorship W3123621629A5044090050 @default.
- W3123621629 hasAuthorship W3123621629A5051959827 @default.
- W3123621629 hasAuthorship W3123621629A5083314813 @default.
- W3123621629 hasConcept C106131492 @default.
- W3123621629 hasConcept C119857082 @default.
- W3123621629 hasConcept C120665830 @default.
- W3123621629 hasConcept C121332964 @default.
- W3123621629 hasConcept C123657996 @default.
- W3123621629 hasConcept C140779682 @default.
- W3123621629 hasConcept C142362112 @default.
- W3123621629 hasConcept C153349607 @default.
- W3123621629 hasConcept C154945302 @default.
- W3123621629 hasConcept C163258240 @default.
- W3123621629 hasConcept C192209626 @default.
- W3123621629 hasConcept C2779803651 @default.
- W3123621629 hasConcept C2780992000 @default.
- W3123621629 hasConcept C31972630 @default.
- W3123621629 hasConcept C41008148 @default.
- W3123621629 hasConcept C62520636 @default.
- W3123621629 hasConcept C76155785 @default.
- W3123621629 hasConcept C94915269 @default.
- W3123621629 hasConceptScore W3123621629C106131492 @default.
- W3123621629 hasConceptScore W3123621629C119857082 @default.
- W3123621629 hasConceptScore W3123621629C120665830 @default.
- W3123621629 hasConceptScore W3123621629C121332964 @default.
- W3123621629 hasConceptScore W3123621629C123657996 @default.
- W3123621629 hasConceptScore W3123621629C140779682 @default.
- W3123621629 hasConceptScore W3123621629C142362112 @default.
- W3123621629 hasConceptScore W3123621629C153349607 @default.
- W3123621629 hasConceptScore W3123621629C154945302 @default.
- W3123621629 hasConceptScore W3123621629C163258240 @default.
- W3123621629 hasConceptScore W3123621629C192209626 @default.
- W3123621629 hasConceptScore W3123621629C2779803651 @default.
- W3123621629 hasConceptScore W3123621629C2780992000 @default.
- W3123621629 hasConceptScore W3123621629C31972630 @default.
- W3123621629 hasConceptScore W3123621629C41008148 @default.
- W3123621629 hasConceptScore W3123621629C62520636 @default.
- W3123621629 hasConceptScore W3123621629C76155785 @default.
- W3123621629 hasConceptScore W3123621629C94915269 @default.
- W3123621629 hasOpenAccess W3123621629 @default.
- W3123621629 hasRelatedWork W12738604 @default.
- W3123621629 hasRelatedWork W134195094 @default.
- W3123621629 hasRelatedWork W1498774938 @default.
- W3123621629 hasRelatedWork W1544782147 @default.
- W3123621629 hasRelatedWork W1644411414 @default.
- W3123621629 hasRelatedWork W2067443455 @default.
- W3123621629 hasRelatedWork W2078069571 @default.
- W3123621629 hasRelatedWork W2113207783 @default.
- W3123621629 hasRelatedWork W2163000150 @default.
- W3123621629 hasRelatedWork W2167105380 @default.
- W3123621629 hasRelatedWork W2402589461 @default.
- W3123621629 hasRelatedWork W2787074649 @default.
- W3123621629 hasRelatedWork W2898658996 @default.
- W3123621629 hasRelatedWork W2990967320 @default.
- W3123621629 hasRelatedWork W2995179471 @default.
- W3123621629 hasRelatedWork W3034202788 @default.
- W3123621629 hasRelatedWork W3084994579 @default.
- W3123621629 hasRelatedWork W3119615817 @default.
- W3123621629 hasRelatedWork W61414934 @default.
- W3123621629 hasRelatedWork W192990226 @default.
- W3123621629 isParatext "false" @default.
- W3123621629 isRetracted "false" @default.
- W3123621629 magId "3123621629" @default.
- W3123621629 workType "article" @default.