FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3166103364> ?p ?o ?g. }

• W3166103364 endingPage "7379" @default.
• W3166103364 startingPage "7367" @default.
• W3166103364 abstract "Compared with traditional convolutions, grouped convolutional neural networks are promising for both model performance and network parameters. However, existing models with the grouped convolution still have parameter redundancy. In this article, concerning the grouped convolution, we propose a sharing grouped convolution structure to reduce parameters. To efficiently eliminate parameter redundancy and improve model performance, we propose a Bayesian sharing framework to transfer the vanilla grouped convolution to be the sharing structure. Intragroup correlation and intergroup importance are introduced into the prior of the parameters. We handle the Maximum Type II likelihood estimation problem of the intragroup correlation and intergroup importance by a group LASSO-type algorithm. The prior mean of the sharing kernels is iteratively updated. Extensive experiments are conducted to demonstrate that on different grouped convolutional neural networks, the proposed sharing grouped convolution structure with the Bayesian sharing framework can reduce parameters and improve prediction accuracy. The proposed sharing framework can reduce parameters up to 64.17%. For ResNeXt-50 with the sharing grouped convolution on ImageNet dataset, network parameters can be reduced by 96.875% in all grouped convolutional layers, and accuracies are improved to 78.86% and 94.54% for top-1 and top-5, respectively." @default.
• W3166103364 created "2021-06-22" @default.
• W3166103364 creator A5002397251 @default.
• W3166103364 creator A5022945347 @default.
• W3166103364 creator A5031741004 @default.
• W3166103364 creator A5043794346 @default.
• W3166103364 creator A5051340429 @default.
• W3166103364 creator A5060117769 @default.
• W3166103364 creator A5074140059 @default.
• W3166103364 creator A5081611476 @default.
• W3166103364 date "2022-12-01" @default.
• W3166103364 modified "2023-10-03" @default.
• W3166103364 title "An Efficient Sharing Grouped Convolution via Bayesian Learning" @default.
• W3166103364 cites W1536680647 @default.
• W3166103364 cites W1887369574 @default.
• W3166103364 cites W2033419225 @default.
• W3166103364 cites W2183341477 @default.
• W3166103364 cites W2194775991 @default.
• W3166103364 cites W2531409750 @default.
• W3166103364 cites W2549139847 @default.
• W3166103364 cites W2778955544 @default.
• W3166103364 cites W2798170643 @default.
• W3166103364 cites W2798429327 @default.
• W3166103364 cites W2884675507 @default.
• W3166103364 cites W2901950140 @default.
• W3166103364 cites W2921569601 @default.
• W3166103364 cites W2944824859 @default.
• W3166103364 cites W2944874836 @default.
• W3166103364 cites W2962851801 @default.
• W3166103364 cites W2962938888 @default.
• W3166103364 cites W2963125010 @default.
• W3166103364 cites W2963363373 @default.
• W3166103364 cites W2963446712 @default.
• W3166103364 cites W2963736842 @default.
• W3166103364 cites W2963993763 @default.
• W3166103364 cites W2965862350 @default.
• W3166103364 cites W2968025890 @default.
• W3166103364 cites W2978081181 @default.
• W3166103364 cites W2981615156 @default.
• W3166103364 cites W2986324132 @default.
• W3166103364 cites W2998543159 @default.
• W3166103364 cites W3005707914 @default.
• W3166103364 cites W3006098440 @default.
• W3166103364 cites W3007902335 @default.
• W3166103364 cites W3021962012 @default.
• W3166103364 cites W3034950896 @default.
• W3166103364 cites W3072011699 @default.
• W3166103364 cites W3095523882 @default.
• W3166103364 cites W3101086546 @default.
• W3166103364 cites W3107401339 @default.
• W3166103364 cites W3111269108 @default.
• W3166103364 cites W4250482878 @default.
• W3166103364 cites W4250589301 @default.
• W3166103364 doi "https://doi.org/10.1109/tnnls.2021.3084900" @default.
• W3166103364 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34111010" @default.
• W3166103364 hasPublicationYear "2022" @default.
• W3166103364 type Work @default.
• W3166103364 sameAs 3166103364 @default.
• W3166103364 citedByCount "9" @default.
• W3166103364 countsByYear W31661033642021 @default.
• W3166103364 countsByYear W31661033642022 @default.
• W3166103364 countsByYear W31661033642023 @default.
• W3166103364 crossrefType "journal-article" @default.
• W3166103364 hasAuthorship W3166103364A5002397251 @default.
• W3166103364 hasAuthorship W3166103364A5022945347 @default.
• W3166103364 hasAuthorship W3166103364A5031741004 @default.
• W3166103364 hasAuthorship W3166103364A5043794346 @default.
• W3166103364 hasAuthorship W3166103364A5051340429 @default.
• W3166103364 hasAuthorship W3166103364A5060117769 @default.
• W3166103364 hasAuthorship W3166103364A5074140059 @default.
• W3166103364 hasAuthorship W3166103364A5081611476 @default.
• W3166103364 hasConcept C107673813 @default.
• W3166103364 hasConcept C111919701 @default.
• W3166103364 hasConcept C11413529 @default.
• W3166103364 hasConcept C119857082 @default.
• W3166103364 hasConcept C152124472 @default.
• W3166103364 hasConcept C153180895 @default.
• W3166103364 hasConcept C154945302 @default.
• W3166103364 hasConcept C41008148 @default.
• W3166103364 hasConcept C45347329 @default.
• W3166103364 hasConcept C50644808 @default.
• W3166103364 hasConcept C81363708 @default.
• W3166103364 hasConceptScore W3166103364C107673813 @default.
• W3166103364 hasConceptScore W3166103364C111919701 @default.
• W3166103364 hasConceptScore W3166103364C11413529 @default.
• W3166103364 hasConceptScore W3166103364C119857082 @default.
• W3166103364 hasConceptScore W3166103364C152124472 @default.
• W3166103364 hasConceptScore W3166103364C153180895 @default.
• W3166103364 hasConceptScore W3166103364C154945302 @default.
• W3166103364 hasConceptScore W3166103364C41008148 @default.
• W3166103364 hasConceptScore W3166103364C45347329 @default.
• W3166103364 hasConceptScore W3166103364C50644808 @default.
• W3166103364 hasConceptScore W3166103364C81363708 @default.
• W3166103364 hasFunder F4320322769 @default.
• W3166103364 hasFunder F4320335777 @default.
• W3166103364 hasIssue "12" @default.
• W3166103364 hasLocation W31661033641 @default.
• W3166103364 hasLocation W31661033642 @default.

• next