Matches in SemOpenAlex for { <https://semopenalex.org/work/W2945783909> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W2945783909 abstract "Artificial neural networks (ANNs) have demonstrated significant promise while implementing recognition and classification applications. The implementation of pre-trained ANNs on embedded systems requires representation of data and design parameters in low-precision fixed-point formats; which often requires retraining of the network. For such implementations, the multiply-accumulate operation is the main reason for resultant high resource and energy requirements. To address these challenges, we present Rox-ANN, a design methodology for implementing ANNs using processing elements (PEs) designed with low-precision fixed-point numbers and high performance and reduced-area approximate multipliers on FPGAs. The trained design parameters of the ANN are analyzed and clustered to optimize the total number of approximate multipliers required in the design. With our methodology, we achieve insignificant loss in application accuracy. We evaluated the design using a LeNet based implementation of the MNIST digit recognition application. The results show a 65.6%, 55.1% and 18.9% reduction in area, energy consumption and latency for a PE using 8-bit precision weights and activations and approximate arithmetic units, when compared to 16-bit full precision, accurate arithmetic PEs." @default.
- W2945783909 created "2019-05-29" @default.
- W2945783909 creator A5028792027 @default.
- W2945783909 creator A5044224621 @default.
- W2945783909 creator A5052267211 @default.
- W2945783909 creator A5076463582 @default.
- W2945783909 date "2019-05-13" @default.
- W2945783909 modified "2023-10-03" @default.
- W2945783909 title "Design Methodology for Embedded Approximate Artificial Neural Networks" @default.
- W2945783909 cites W1974078116 @default.
- W2945783909 cites W1998917233 @default.
- W2945783909 cites W2013305145 @default.
- W2945783909 cites W2051831389 @default.
- W2945783909 cites W2094756095 @default.
- W2945783909 cites W2147768505 @default.
- W2945783909 cites W2289252105 @default.
- W2945783909 cites W2299244345 @default.
- W2945783909 cites W2396572963 @default.
- W2945783909 cites W2616014673 @default.
- W2945783909 cites W2725615981 @default.
- W2945783909 cites W2808908725 @default.
- W2945783909 cites W3001665736 @default.
- W2945783909 cites W4212788319 @default.
- W2945783909 doi "https://doi.org/10.1145/3299874.3319490" @default.
- W2945783909 hasPublicationYear "2019" @default.
- W2945783909 type Work @default.
- W2945783909 sameAs 2945783909 @default.
- W2945783909 citedByCount "14" @default.
- W2945783909 countsByYear W29457839092020 @default.
- W2945783909 countsByYear W29457839092021 @default.
- W2945783909 countsByYear W29457839092022 @default.
- W2945783909 crossrefType "proceedings-article" @default.
- W2945783909 hasAuthorship W2945783909A5028792027 @default.
- W2945783909 hasAuthorship W2945783909A5044224621 @default.
- W2945783909 hasAuthorship W2945783909A5052267211 @default.
- W2945783909 hasAuthorship W2945783909A5076463582 @default.
- W2945783909 hasConcept C113775141 @default.
- W2945783909 hasConcept C119599485 @default.
- W2945783909 hasConcept C124584101 @default.
- W2945783909 hasConcept C127413603 @default.
- W2945783909 hasConcept C134835016 @default.
- W2945783909 hasConcept C139719470 @default.
- W2945783909 hasConcept C154945302 @default.
- W2945783909 hasConcept C162324750 @default.
- W2945783909 hasConcept C190502265 @default.
- W2945783909 hasConcept C199360897 @default.
- W2945783909 hasConcept C26713055 @default.
- W2945783909 hasConcept C2780165032 @default.
- W2945783909 hasConcept C41008148 @default.
- W2945783909 hasConcept C42935608 @default.
- W2945783909 hasConcept C50644808 @default.
- W2945783909 hasConcept C76155785 @default.
- W2945783909 hasConcept C82876162 @default.
- W2945783909 hasConcept C9390403 @default.
- W2945783909 hasConceptScore W2945783909C113775141 @default.
- W2945783909 hasConceptScore W2945783909C119599485 @default.
- W2945783909 hasConceptScore W2945783909C124584101 @default.
- W2945783909 hasConceptScore W2945783909C127413603 @default.
- W2945783909 hasConceptScore W2945783909C134835016 @default.
- W2945783909 hasConceptScore W2945783909C139719470 @default.
- W2945783909 hasConceptScore W2945783909C154945302 @default.
- W2945783909 hasConceptScore W2945783909C162324750 @default.
- W2945783909 hasConceptScore W2945783909C190502265 @default.
- W2945783909 hasConceptScore W2945783909C199360897 @default.
- W2945783909 hasConceptScore W2945783909C26713055 @default.
- W2945783909 hasConceptScore W2945783909C2780165032 @default.
- W2945783909 hasConceptScore W2945783909C41008148 @default.
- W2945783909 hasConceptScore W2945783909C42935608 @default.
- W2945783909 hasConceptScore W2945783909C50644808 @default.
- W2945783909 hasConceptScore W2945783909C76155785 @default.
- W2945783909 hasConceptScore W2945783909C82876162 @default.
- W2945783909 hasConceptScore W2945783909C9390403 @default.
- W2945783909 hasLocation W29457839091 @default.
- W2945783909 hasOpenAccess W2945783909 @default.
- W2945783909 hasPrimaryLocation W29457839091 @default.
- W2945783909 hasRelatedWork W2130748623 @default.
- W2945783909 hasRelatedWork W2144044430 @default.
- W2945783909 hasRelatedWork W2156094073 @default.
- W2945783909 hasRelatedWork W2160284536 @default.
- W2945783909 hasRelatedWork W2256896364 @default.
- W2945783909 hasRelatedWork W2378665840 @default.
- W2945783909 hasRelatedWork W2391017997 @default.
- W2945783909 hasRelatedWork W2953292182 @default.
- W2945783909 hasRelatedWork W3013923459 @default.
- W2945783909 hasRelatedWork W2547610195 @default.
- W2945783909 isParatext "false" @default.
- W2945783909 isRetracted "false" @default.
- W2945783909 magId "2945783909" @default.
- W2945783909 workType "article" @default.