Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288374925> ?p ?o ?g. }
Showing items 1 to 33 of
33
with 100 items per page.
- W4288374925 abstract "In recent years, there has been increasing demand for automatic architecture search in deep learning. Numerous approaches have been proposed and led to state-of-the-art results in various applications, including image classification and language modeling. In this paper, we propose a novel way of architecture search by means of weighted networks (WeNet), which consist of a number of networks, with each assigned a weight. These weights are updated with back-propagation to reflect the importance of different networks. Such weighted networks bear similarity to mixture of experts. We conduct experiments on Penn Treebank and WikiText-2. We show that the proposed WeNet can find recurrent architectures which result in state-of-the-art performance." @default.
- W4288374925 created "2022-07-29" @default.
- W4288374925 creator A5001906708 @default.
- W4288374925 creator A5011349177 @default.
- W4288374925 date "2019-04-07" @default.
- W4288374925 modified "2023-09-27" @default.
- W4288374925 title "WeNet: Weighted Networks for Recurrent Network Architecture Search" @default.
- W4288374925 doi "https://doi.org/10.48550/arxiv.1904.03819" @default.
- W4288374925 hasPublicationYear "2019" @default.
- W4288374925 type Work @default.
- W4288374925 citedByCount "0" @default.
- W4288374925 crossrefType "posted-content" @default.
- W4288374925 hasAuthorship W4288374925A5001906708 @default.
- W4288374925 hasAuthorship W4288374925A5011349177 @default.
- W4288374925 hasBestOaLocation W42883749251 @default.
- W4288374925 hasConcept C123657996 @default.
- W4288374925 hasConcept C166957645 @default.
- W4288374925 hasConcept C193415008 @default.
- W4288374925 hasConcept C205649164 @default.
- W4288374925 hasConcept C31258907 @default.
- W4288374925 hasConcept C41008148 @default.
- W4288374925 hasConceptScore W4288374925C123657996 @default.
- W4288374925 hasConceptScore W4288374925C166957645 @default.
- W4288374925 hasConceptScore W4288374925C193415008 @default.
- W4288374925 hasConceptScore W4288374925C205649164 @default.
- W4288374925 hasConceptScore W4288374925C31258907 @default.
- W4288374925 hasConceptScore W4288374925C41008148 @default.
- W4288374925 hasLocation W42883749251 @default.
- W4288374925 hasOpenAccess W4288374925 @default.
- W4288374925 hasPrimaryLocation W42883749251 @default.
- W4288374925 isParatext "false" @default.
- W4288374925 isRetracted "false" @default.
- W4288374925 workType "article" @default.