Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783156638> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2783156638 abstract "Convolution neural networks (CNNs) have become the state-of-the-art technique for Artificial Intelligence (AI) applications. But they consume resources heavily both in computation and energy. Compressing the parameters to fewer bits is an effective way to reduce data quantity, making it possible to deploy CNNs on mobile, IoT and other resource-limited devices. The existing means need to re-train the entire networks and tune the parameters. This paper presents a convenient method to compress the parameters of trained CNN models without re-training and fine-tuning. Our method reduces the number of bits from 32 to 5-9 under the same networks and on the same dataset, with the loss of accuracy less than 1%. The logic power of a trained LeNet reduces by 70.58% on FPGA evaluation." @default.
- W2783156638 created "2018-01-26" @default.
- W2783156638 creator A5046982010 @default.
- W2783156638 creator A5056168495 @default.
- W2783156638 creator A5062671813 @default.
- W2783156638 creator A5089180350 @default.
- W2783156638 creator A5090493433 @default.
- W2783156638 date "2017-10-01" @default.
- W2783156638 modified "2023-09-25" @default.
- W2783156638 title "A low bit-width parameter representation method for hardware-oriented convolution neural networks" @default.
- W2783156638 cites W1999085092 @default.
- W2783156638 cites W2048266589 @default.
- W2783156638 cites W2106033855 @default.
- W2783156638 cites W2152839228 @default.
- W2783156638 cites W2594836184 @default.
- W2783156638 doi "https://doi.org/10.1109/asicon.2017.8252433" @default.
- W2783156638 hasPublicationYear "2017" @default.
- W2783156638 type Work @default.
- W2783156638 sameAs 2783156638 @default.
- W2783156638 citedByCount "3" @default.
- W2783156638 countsByYear W27831566382019 @default.
- W2783156638 countsByYear W27831566382020 @default.
- W2783156638 countsByYear W27831566382023 @default.
- W2783156638 crossrefType "proceedings-article" @default.
- W2783156638 hasAuthorship W2783156638A5046982010 @default.
- W2783156638 hasAuthorship W2783156638A5056168495 @default.
- W2783156638 hasAuthorship W2783156638A5062671813 @default.
- W2783156638 hasAuthorship W2783156638A5089180350 @default.
- W2783156638 hasAuthorship W2783156638A5090493433 @default.
- W2783156638 hasConcept C111919701 @default.
- W2783156638 hasConcept C113775141 @default.
- W2783156638 hasConcept C11413529 @default.
- W2783156638 hasConcept C149635348 @default.
- W2783156638 hasConcept C154945302 @default.
- W2783156638 hasConcept C17744445 @default.
- W2783156638 hasConcept C186967261 @default.
- W2783156638 hasConcept C199539241 @default.
- W2783156638 hasConcept C2776359362 @default.
- W2783156638 hasConcept C41008148 @default.
- W2783156638 hasConcept C42935608 @default.
- W2783156638 hasConcept C45347329 @default.
- W2783156638 hasConcept C45374587 @default.
- W2783156638 hasConcept C50644808 @default.
- W2783156638 hasConcept C81363708 @default.
- W2783156638 hasConcept C9390403 @default.
- W2783156638 hasConcept C94625758 @default.
- W2783156638 hasConceptScore W2783156638C111919701 @default.
- W2783156638 hasConceptScore W2783156638C113775141 @default.
- W2783156638 hasConceptScore W2783156638C11413529 @default.
- W2783156638 hasConceptScore W2783156638C149635348 @default.
- W2783156638 hasConceptScore W2783156638C154945302 @default.
- W2783156638 hasConceptScore W2783156638C17744445 @default.
- W2783156638 hasConceptScore W2783156638C186967261 @default.
- W2783156638 hasConceptScore W2783156638C199539241 @default.
- W2783156638 hasConceptScore W2783156638C2776359362 @default.
- W2783156638 hasConceptScore W2783156638C41008148 @default.
- W2783156638 hasConceptScore W2783156638C42935608 @default.
- W2783156638 hasConceptScore W2783156638C45347329 @default.
- W2783156638 hasConceptScore W2783156638C45374587 @default.
- W2783156638 hasConceptScore W2783156638C50644808 @default.
- W2783156638 hasConceptScore W2783156638C81363708 @default.
- W2783156638 hasConceptScore W2783156638C9390403 @default.
- W2783156638 hasConceptScore W2783156638C94625758 @default.
- W2783156638 hasLocation W27831566381 @default.
- W2783156638 hasOpenAccess W2783156638 @default.
- W2783156638 hasPrimaryLocation W27831566381 @default.
- W2783156638 hasRelatedWork W2067951144 @default.
- W2783156638 hasRelatedWork W2068395868 @default.
- W2783156638 hasRelatedWork W2233116163 @default.
- W2783156638 hasRelatedWork W2561485601 @default.
- W2783156638 hasRelatedWork W2796942851 @default.
- W2783156638 hasRelatedWork W2902363683 @default.
- W2783156638 hasRelatedWork W2949961122 @default.
- W2783156638 hasRelatedWork W2952735458 @default.
- W2783156638 hasRelatedWork W4312417841 @default.
- W2783156638 hasRelatedWork W4319662858 @default.
- W2783156638 isParatext "false" @default.
- W2783156638 isRetracted "false" @default.
- W2783156638 magId "2783156638" @default.
- W2783156638 workType "article" @default.