Matches in SemOpenAlex for { <https://semopenalex.org/work/W2924493694> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W2924493694 endingPage "032024" @default.
- W2924493694 startingPage "032024" @default.
- W2924493694 abstract "With the advancements of circuit design and manufacturing processes of hardware, Deep Neural Networks (DNN) have been developed to be deeper and larger to achieve remarkable results in many aspects. Despite this, problems such as excessive memory usage and severe resource consumption are still unavoidable. We proposed a symmetric quantization scheme, which quantizes the weights from a certain kernel to be symmetric to zero along with a scaling factor and bias term in order to make full use of the whole quantization space and reduce the multiplication operations when convolution happens. We also came up with an Approximate Batch Normalization (ABN) method, which simplifies the traditional BN to linear operations lightly. Besides, activations and gradients are quantized in our scheme with fine distinctions, so that most floating-point calculations can be converted into fixed-point calculations and bit-shift operations in both training and inference phase. Our proposed scheme greatly alleviates the problems mentioned above theoretically with fewer computations and smaller storage theoretically. We carried out various experiments to prove its feasibility." @default.
- W2924493694 created "2019-04-01" @default.
- W2924493694 creator A5050022243 @default.
- W2924493694 creator A5074882411 @default.
- W2924493694 creator A5090679516 @default.
- W2924493694 date "2019-03-01" @default.
- W2924493694 modified "2023-09-25" @default.
- W2924493694 title "Symmetric Quantization Scheme and Approximate Batch Normalization for Quantized Neural Networks Training" @default.
- W2924493694 cites W2058641082 @default.
- W2924493694 doi "https://doi.org/10.1088/1742-6596/1176/3/032024" @default.
- W2924493694 hasPublicationYear "2019" @default.
- W2924493694 type Work @default.
- W2924493694 sameAs 2924493694 @default.
- W2924493694 citedByCount "1" @default.
- W2924493694 countsByYear W29244936942019 @default.
- W2924493694 crossrefType "journal-article" @default.
- W2924493694 hasAuthorship W2924493694A5050022243 @default.
- W2924493694 hasAuthorship W2924493694A5074882411 @default.
- W2924493694 hasAuthorship W2924493694A5090679516 @default.
- W2924493694 hasBestOaLocation W29244936941 @default.
- W2924493694 hasConcept C113775141 @default.
- W2924493694 hasConcept C11413529 @default.
- W2924493694 hasConcept C114614502 @default.
- W2924493694 hasConcept C118615104 @default.
- W2924493694 hasConcept C136886441 @default.
- W2924493694 hasConcept C144024400 @default.
- W2924493694 hasConcept C154945302 @default.
- W2924493694 hasConcept C19165224 @default.
- W2924493694 hasConcept C2524010 @default.
- W2924493694 hasConcept C2776214188 @default.
- W2924493694 hasConcept C2780595030 @default.
- W2924493694 hasConcept C28855332 @default.
- W2924493694 hasConcept C2984842247 @default.
- W2924493694 hasConcept C33923547 @default.
- W2924493694 hasConcept C41008148 @default.
- W2924493694 hasConcept C45374587 @default.
- W2924493694 hasConcept C50644808 @default.
- W2924493694 hasConcept C74193536 @default.
- W2924493694 hasConcept C99844830 @default.
- W2924493694 hasConceptScore W2924493694C113775141 @default.
- W2924493694 hasConceptScore W2924493694C11413529 @default.
- W2924493694 hasConceptScore W2924493694C114614502 @default.
- W2924493694 hasConceptScore W2924493694C118615104 @default.
- W2924493694 hasConceptScore W2924493694C136886441 @default.
- W2924493694 hasConceptScore W2924493694C144024400 @default.
- W2924493694 hasConceptScore W2924493694C154945302 @default.
- W2924493694 hasConceptScore W2924493694C19165224 @default.
- W2924493694 hasConceptScore W2924493694C2524010 @default.
- W2924493694 hasConceptScore W2924493694C2776214188 @default.
- W2924493694 hasConceptScore W2924493694C2780595030 @default.
- W2924493694 hasConceptScore W2924493694C28855332 @default.
- W2924493694 hasConceptScore W2924493694C2984842247 @default.
- W2924493694 hasConceptScore W2924493694C33923547 @default.
- W2924493694 hasConceptScore W2924493694C41008148 @default.
- W2924493694 hasConceptScore W2924493694C45374587 @default.
- W2924493694 hasConceptScore W2924493694C50644808 @default.
- W2924493694 hasConceptScore W2924493694C74193536 @default.
- W2924493694 hasConceptScore W2924493694C99844830 @default.
- W2924493694 hasLocation W29244936941 @default.
- W2924493694 hasOpenAccess W2924493694 @default.
- W2924493694 hasPrimaryLocation W29244936941 @default.
- W2924493694 hasRelatedWork W2777406049 @default.
- W2924493694 hasRelatedWork W2803935332 @default.
- W2924493694 hasRelatedWork W2891335318 @default.
- W2924493694 hasRelatedWork W2898755250 @default.
- W2924493694 hasRelatedWork W3082154786 @default.
- W2924493694 hasRelatedWork W3099092507 @default.
- W2924493694 hasRelatedWork W3217204427 @default.
- W2924493694 hasRelatedWork W4213041209 @default.
- W2924493694 hasRelatedWork W4310823283 @default.
- W2924493694 hasRelatedWork W4378468568 @default.
- W2924493694 hasVolume "1176" @default.
- W2924493694 isParatext "false" @default.
- W2924493694 isRetracted "false" @default.
- W2924493694 magId "2924493694" @default.
- W2924493694 workType "article" @default.