FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3044853878> ?p ?o ?g. }

• W3044853878 abstract "In this paper, we propose an efficient NAS algorithm for generating task-specific models that are competitive under multiple competing objectives. It comprises of two surrogates, one at the architecture level to improve sample efficiency and one at the weights level, through a supernet, to improve gradient descent training efficiency. On standard benchmark datasets (C10, C100, ImageNet), the resulting models, dubbed NSGANetV2, either match or outperform models from existing approaches with the search being orders of magnitude more sample efficient. Furthermore, we demonstrate the effectiveness and versatility of the proposed method on six diverse non-standard datasets, e.g. STL-10, Flowers102, Oxford Pets, FGVC Aircrafts etc. In all cases, NSGANetV2s improve the state-of-the-art (under mobile setting), suggesting that NAS can be a viable alternative to conventional transfer learning approaches in handling diverse scenarios such as small-scale or fine-grained datasets. Code is available at https://github.com/mikelzc1990/nsganetv2" @default.
• W3044853878 created "2020-07-29" @default.
• W3044853878 creator A5004837138 @default.
• W3044853878 creator A5027687139 @default.
• W3044853878 creator A5031717929 @default.
• W3044853878 creator A5034853476 @default.
• W3044853878 creator A5088394271 @default.
• W3044853878 date "2020-07-20" @default.
• W3044853878 modified "2023-10-14" @default.
• W3044853878 title "NSGANetV2: Evolutionary Multi-Objective Surrogate-Assisted Neural Architecture Search" @default.
• W3044853878 cites W1558919105 @default.
• W3044853878 cites W1846799578 @default.
• W3044853878 cites W1977295328 @default.
• W3044853878 cites W1985514943 @default.
• W3044853878 cites W2031264011 @default.
• W3044853878 cites W2047643928 @default.
• W3044853878 cites W2117539524 @default.
• W3044853878 cites W2118858186 @default.
• W3044853878 cites W2126105956 @default.
• W3044853878 cites W2533598788 @default.
• W3044853878 cites W2594586811 @default.
• W3044853878 cites W2808938483 @default.
• W3044853878 cites W2810075754 @default.
• W3044853878 cites W2888429796 @default.
• W3044853878 cites W2895171208 @default.
• W3044853878 cites W2917028965 @default.
• W3044853878 cites W2944779197 @default.
• W3044853878 cites W2953604046 @default.
• W3044853878 cites W2954234207 @default.
• W3044853878 cites W2954826798 @default.
• W3044853878 cites W2955425717 @default.
• W3044853878 cites W2962746461 @default.
• W3044853878 cites W2963137684 @default.
• W3044853878 cites W2963163009 @default.
• W3044853878 cites W2963821229 @default.
• W3044853878 cites W2963918968 @default.
• W3044853878 cites W2964081807 @default.
• W3044853878 cites W2964259004 @default.
• W3044853878 cites W2964294659 @default.
• W3044853878 cites W2964331719 @default.
• W3044853878 cites W2965658867 @default.
• W3044853878 cites W2967733054 @default.
• W3044853878 cites W2971107871 @default.
• W3044853878 cites W2978426779 @default.
• W3044853878 cites W2980137827 @default.
• W3044853878 cites W2981985696 @default.
• W3044853878 cites W2982083293 @default.
• W3044853878 cites W2990721113 @default.
• W3044853878 cites W2994749257 @default.
• W3044853878 cites W2995727387 @default.
• W3044853878 cites W3022419825 @default.
• W3044853878 cites W3034230425 @default.
• W3044853878 cites W3118608800 @default.
• W3044853878 cites W3127664874 @default.
• W3044853878 doi "https://doi.org/10.48550/arxiv.2007.10396" @default.
• W3044853878 hasPublicationYear "2020" @default.
• W3044853878 type Work @default.
• W3044853878 sameAs 3044853878 @default.
• W3044853878 citedByCount "3" @default.
• W3044853878 countsByYear W30448538782020 @default.
• W3044853878 countsByYear W30448538782021 @default.
• W3044853878 crossrefType "posted-content" @default.
• W3044853878 hasAuthorship W3044853878A5004837138 @default.
• W3044853878 hasAuthorship W3044853878A5027687139 @default.
• W3044853878 hasAuthorship W3044853878A5031717929 @default.
• W3044853878 hasAuthorship W3044853878A5034853476 @default.
• W3044853878 hasAuthorship W3044853878A5088394271 @default.
• W3044853878 hasBestOaLocation W30448538781 @default.
• W3044853878 hasConcept C119857082 @default.
• W3044853878 hasConcept C123657996 @default.
• W3044853878 hasConcept C124101348 @default.
• W3044853878 hasConcept C127413603 @default.
• W3044853878 hasConcept C13280743 @default.
• W3044853878 hasConcept C142362112 @default.
• W3044853878 hasConcept C150899416 @default.
• W3044853878 hasConcept C153349607 @default.
• W3044853878 hasConcept C154945302 @default.
• W3044853878 hasConcept C177264268 @default.
• W3044853878 hasConcept C185592680 @default.
• W3044853878 hasConcept C185798385 @default.
• W3044853878 hasConcept C198531522 @default.
• W3044853878 hasConcept C199360897 @default.
• W3044853878 hasConcept C201995342 @default.
• W3044853878 hasConcept C205649164 @default.
• W3044853878 hasConcept C2776760102 @default.
• W3044853878 hasConcept C2780451532 @default.
• W3044853878 hasConcept C41008148 @default.
• W3044853878 hasConcept C43617362 @default.
• W3044853878 hasConceptScore W3044853878C119857082 @default.
• W3044853878 hasConceptScore W3044853878C123657996 @default.
• W3044853878 hasConceptScore W3044853878C124101348 @default.
• W3044853878 hasConceptScore W3044853878C127413603 @default.
• W3044853878 hasConceptScore W3044853878C13280743 @default.
• W3044853878 hasConceptScore W3044853878C142362112 @default.
• W3044853878 hasConceptScore W3044853878C150899416 @default.
• W3044853878 hasConceptScore W3044853878C153349607 @default.
• W3044853878 hasConceptScore W3044853878C154945302 @default.
• W3044853878 hasConceptScore W3044853878C177264268 @default.
• W3044853878 hasConceptScore W3044853878C185592680 @default.
• W3044853878 hasConceptScore W3044853878C185798385 @default.

• next