FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2986308587> ?p ?o ?g. }

• W2986308587 abstract "Modern convolutional neural networks architectures are very resource intensive which limits the possibilities for their wide practical application. We propose a convolutional neural network architecture in which the neural network is divided into hardware and software parts to increase performance and reduce the cost of implementation resources. We also propose to use the residue number system in the hardware part to implement the convolutional layer of the neural network for resource costs reducing. A numerical method for quantizing the filters coefficients of a convolutional network layer is proposed to minimize the influence of quantization noise on the calculation result in the residue number system and determine the bit-width of the filters coefficients. This method is based on scaling the coefficients by a fixed number of bits and rounding up and down. The operations used make it possible to reduce resources in hardware implementation due to the simplifying of their execution. All calculations in the convolutional layer are performed on numbers in a fixed-point format. Software simulations using Matlab 2017b showed that convolutional neural network with a minimum number of layers can be quickly and successfully trained. Hardware implementation using the field-programmable gate array Kintex7 xc7k70tfbg484-2 showed that the use of residue number system in the convolutional layer of the neural network reduces the hardware costs by 32.6% compared with the traditional approach based on the two’s complement representation. The research results can be applied to create effective video surveillance systems, for recognizing handwriting, individuals, objects and terrain." @default.
• W2986308587 created "2019-11-22" @default.
• W2986308587 creator A5004597011 @default.
• W2986308587 creator A5018867267 @default.
• W2986308587 creator A5019972460 @default.
• W2986308587 creator A5032223985 @default.
• W2986308587 creator A5090768485 @default.
• W2986308587 date "2019-10-01" @default.
• W2986308587 modified "2023-10-15" @default.
• W2986308587 title "Hardware implementation of a convolutional neural network using calculations in the residue number system" @default.
• W2986308587 cites W1505484906 @default.
• W2986308587 cites W1672699986 @default.
• W2986308587 cites W2094799249 @default.
• W2986308587 cites W2097117768 @default.
• W2986308587 cites W2112796928 @default.
• W2986308587 cites W2137898528 @default.
• W2986308587 cites W2164556899 @default.
• W2986308587 cites W2529865900 @default.
• W2986308587 cites W2545238695 @default.
• W2986308587 cites W2564933552 @default.
• W2986308587 cites W2601707599 @default.
• W2986308587 cites W2618530766 @default.
• W2986308587 cites W2755125693 @default.
• W2986308587 cites W2760772103 @default.
• W2986308587 cites W2762624795 @default.
• W2986308587 cites W2768941608 @default.
• W2986308587 cites W2786340847 @default.
• W2986308587 cites W2799995256 @default.
• W2986308587 cites W2801748224 @default.
• W2986308587 cites W2883200793 @default.
• W2986308587 cites W2889564963 @default.
• W2986308587 cites W2906258714 @default.
• W2986308587 cites W2911566884 @default.
• W2986308587 cites W2962987932 @default.
• W2986308587 cites W4229484196 @default.
• W2986308587 doi "https://doi.org/10.18287/2412-6179-2019-43-5-857-868" @default.
• W2986308587 hasPublicationYear "2019" @default.
• W2986308587 type Work @default.
• W2986308587 sameAs 2986308587 @default.
• W2986308587 citedByCount "6" @default.
• W2986308587 countsByYear W29863085872020 @default.
• W2986308587 countsByYear W29863085872022 @default.
• W2986308587 crossrefType "journal-article" @default.
• W2986308587 hasAuthorship W2986308587A5004597011 @default.
• W2986308587 hasAuthorship W2986308587A5018867267 @default.
• W2986308587 hasAuthorship W2986308587A5019972460 @default.
• W2986308587 hasAuthorship W2986308587A5032223985 @default.
• W2986308587 hasAuthorship W2986308587A5090768485 @default.
• W2986308587 hasBestOaLocation W29863085871 @default.
• W2986308587 hasConcept C108583219 @default.
• W2986308587 hasConcept C111919701 @default.
• W2986308587 hasConcept C113775141 @default.
• W2986308587 hasConcept C11413529 @default.
• W2986308587 hasConcept C136625980 @default.
• W2986308587 hasConcept C154945302 @default.
• W2986308587 hasConcept C173608175 @default.
• W2986308587 hasConcept C199360897 @default.
• W2986308587 hasConcept C2777904410 @default.
• W2986308587 hasConcept C41008148 @default.
• W2986308587 hasConcept C50644808 @default.
• W2986308587 hasConcept C79403827 @default.
• W2986308587 hasConcept C81363708 @default.
• W2986308587 hasConcept C84211073 @default.
• W2986308587 hasConcept C9390403 @default.
• W2986308587 hasConceptScore W2986308587C108583219 @default.
• W2986308587 hasConceptScore W2986308587C111919701 @default.
• W2986308587 hasConceptScore W2986308587C113775141 @default.
• W2986308587 hasConceptScore W2986308587C11413529 @default.
• W2986308587 hasConceptScore W2986308587C136625980 @default.
• W2986308587 hasConceptScore W2986308587C154945302 @default.
• W2986308587 hasConceptScore W2986308587C173608175 @default.
• W2986308587 hasConceptScore W2986308587C199360897 @default.
• W2986308587 hasConceptScore W2986308587C2777904410 @default.
• W2986308587 hasConceptScore W2986308587C41008148 @default.
• W2986308587 hasConceptScore W2986308587C50644808 @default.
• W2986308587 hasConceptScore W2986308587C79403827 @default.
• W2986308587 hasConceptScore W2986308587C81363708 @default.
• W2986308587 hasConceptScore W2986308587C84211073 @default.
• W2986308587 hasConceptScore W2986308587C9390403 @default.
• W2986308587 hasFunder F4320321079 @default.
• W2986308587 hasFunder F4320321912 @default.
• W2986308587 hasIssue "5" @default.
• W2986308587 hasLocation W29863085871 @default.
• W2986308587 hasLocation W29863085872 @default.
• W2986308587 hasOpenAccess W2986308587 @default.
• W2986308587 hasPrimaryLocation W29863085871 @default.
• W2986308587 hasRelatedWork W1981434342 @default.
• W2986308587 hasRelatedWork W2106652828 @default.
• W2986308587 hasRelatedWork W2392910503 @default.
• W2986308587 hasRelatedWork W2541249978 @default.
• W2986308587 hasRelatedWork W2889797931 @default.
• W2986308587 hasRelatedWork W2890840567 @default.
• W2986308587 hasRelatedWork W2947629474 @default.
• W2986308587 hasRelatedWork W2986308587 @default.
• W2986308587 hasRelatedWork W3153800852 @default.
• W2986308587 hasRelatedWork W3185228140 @default.
• W2986308587 hasVolume "43" @default.
• W2986308587 isParatext "false" @default.
• W2986308587 isRetracted "false" @default.
• W2986308587 magId "2986308587" @default.

• next