FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3160736633> ?p ?o ?g. }

• W3160736633 abstract "Transfer learning can boost the performance on the targettask by leveraging the knowledge of the source domain. Recent worksin neural architecture search (NAS), especially one-shot NAS, can aidtransfer learning by establishing sufficient network search space. How-ever, existing NAS methods tend to approximate huge search spaces byexplicitly building giant super-networks with multiple sub-paths, anddiscard super-network weights after a child structure is found. Both thecharacteristics of existing approaches causes repetitive network trainingon source tasks in transfer learning. To remedy the above issues, we re-duce the super-network size by randomly dropping connection betweennetwork blocks while embedding a larger search space. Moreover, wereuse super-network weights to avoid redundant training by proposinga novel framework consisting of two modules, the neural architecturesearch module for architecture transfer and the neural weight searchmodule for weight transfer. These two modules conduct search on thetarget task based on a reduced super-networks, so we only need to trainonce on the source task. We experiment our framework on both MS-COCO and CUB-200 for the object detection and fine-grained imageclassification tasks, and show promising improvements with onlyO(CN)super-network complexity." @default.
• W3160736633 created "2021-05-24" @default.
• W3160736633 creator A5054033403 @default.
• W3160736633 creator A5069611975 @default.
• W3160736633 creator A5076466251 @default.
• W3160736633 date "2021-05-19" @default.
• W3160736633 modified "2023-09-24" @default.
• W3160736633 title "Efficient Transfer Learning via Joint Adaptation of Network Architecture and Weight" @default.
• W3160736633 cites W1565327149 @default.
• W3160736633 cites W1731081199 @default.
• W3160736633 cites W1797268635 @default.
• W3160736633 cites W1821462560 @default.
• W3160736633 cites W1861492603 @default.
• W3160736633 cites W2122838776 @default.
• W3160736633 cites W2125918842 @default.
• W3160736633 cites W2138857742 @default.
• W3160736633 cites W2165698076 @default.
• W3160736633 cites W2214409633 @default.
• W3160736633 cites W2331143823 @default.
• W3160736633 cites W2549139847 @default.
• W3160736633 cites W2553303224 @default.
• W3160736633 cites W2556967412 @default.
• W3160736633 cites W2557728737 @default.
• W3160736633 cites W2565639579 @default.
• W3160736633 cites W2588646734 @default.
• W3160736633 cites W2593744649 @default.
• W3160736633 cites W2593768305 @default.
• W3160736633 cites W2594529350 @default.
• W3160736633 cites W2613718673 @default.
• W3160736633 cites W2616287544 @default.
• W3160736633 cites W2750432752 @default.
• W3160736633 cites W2771083582 @default.
• W3160736633 cites W2772955562 @default.
• W3160736633 cites W2782417188 @default.
• W3160736633 cites W2783000019 @default.
• W3160736633 cites W2785366763 @default.
• W3160736633 cites W2785430118 @default.
• W3160736633 cites W2798381792 @default.
• W3160736633 cites W2799269579 @default.
• W3160736633 cites W2804935296 @default.
• W3160736633 cites W2807931652 @default.
• W3160736633 cites W2885820039 @default.
• W3160736633 cites W2887280559 @default.
• W3160736633 cites W2890166761 @default.
• W3160736633 cites W2899771611 @default.
• W3160736633 cites W2946232168 @default.
• W3160736633 cites W2949650786 @default.
• W3160736633 cites W2951670162 @default.
• W3160736633 cites W2951886768 @default.
• W3160736633 cites W2955425717 @default.
• W3160736633 cites W2963091558 @default.
• W3160736633 cites W2963150697 @default.
• W3160736633 cites W2963266682 @default.
• W3160736633 cites W2963420686 @default.
• W3160736633 cites W2963826681 @default.
• W3160736633 cites W2963855133 @default.
• W3160736633 cites W2963918968 @default.
• W3160736633 cites W2964081807 @default.
• W3160736633 cites W2964093967 @default.
• W3160736633 cites W2964259004 @default.
• W3160736633 cites W2967733054 @default.
• W3160736633 cites W2996407667 @default.
• W3160736633 cites W14333344 @default.
• W3160736633 doi "https://doi.org/10.48550/arxiv.2105.08994" @default.
• W3160736633 hasPublicationYear "2021" @default.
• W3160736633 type Work @default.
• W3160736633 sameAs 3160736633 @default.
• W3160736633 citedByCount "0" @default.
• W3160736633 crossrefType "posted-content" @default.
• W3160736633 hasAuthorship W3160736633A5054033403 @default.
• W3160736633 hasAuthorship W3160736633A5069611975 @default.
• W3160736633 hasAuthorship W3160736633A5076466251 @default.
• W3160736633 hasBestOaLocation W31607366331 @default.
• W3160736633 hasConcept C111919701 @default.
• W3160736633 hasConcept C119857082 @default.
• W3160736633 hasConcept C120314980 @default.
• W3160736633 hasConcept C120665830 @default.
• W3160736633 hasConcept C121332964 @default.
• W3160736633 hasConcept C123657996 @default.
• W3160736633 hasConcept C127413603 @default.
• W3160736633 hasConcept C134306372 @default.
• W3160736633 hasConcept C139807058 @default.
• W3160736633 hasConcept C142362112 @default.
• W3160736633 hasConcept C150899416 @default.
• W3160736633 hasConcept C153349607 @default.
• W3160736633 hasConcept C154945302 @default.
• W3160736633 hasConcept C173608175 @default.
• W3160736633 hasConcept C193415008 @default.
• W3160736633 hasConcept C201995342 @default.
• W3160736633 hasConcept C2776175482 @default.
• W3160736633 hasConcept C2778572836 @default.
• W3160736633 hasConcept C2780451532 @default.
• W3160736633 hasConcept C2781238097 @default.
• W3160736633 hasConcept C31258907 @default.
• W3160736633 hasConcept C33923547 @default.
• W3160736633 hasConcept C36503486 @default.
• W3160736633 hasConcept C41008148 @default.
• W3160736633 hasConcept C41608201 @default.
• W3160736633 hasConcept C50644808 @default.
• W3160736633 hasConcept C80444323 @default.

• next