Matches in SemOpenAlex for { <https://semopenalex.org/work/W2798317693> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2798317693 abstract "Convolutional Neural Network (CNN) accelerators are rapidly growing in popularity as a promising solution for deep learning based applications. Though optimizations on computation have been intensively studied, the energy efficiency of such accelerators remains limited by off-chip memory accesses since their energy cost is magnitudes higher than other operations. Minimizing off-chip memory access volume, therefore, is the key to further improving energy efficiency. However, we observed that sticking to minimizing the accesses of one data type as many prior work did cannot fit the varying shapes of convolutional layers in CNNs. Hence, there exists a dilemma of minimizing the accesses of which data type. To overcome the problem, this paper proposed an adaptive layer partitioning and scheduling scheme, called SmartShuttle, to minimize off-chip memory accesses for CNN accelerators. Smartshuttle can adaptively switch among different data reuse schemes and the corresponding tiling factor settings to dynamically match different convolutional layers. Moreover, SmartShuttle thoroughly investigates the impact of data reusability and sparsity on the memory access volume. The experimental results show that SmartShuttle processes the convolutional layers at 434.8 multiply and accumulations (MACs)/DRAM access for VGG16 (batch size = 3), and 526.3 MACs/DRAM access for AlexNet (batch size = 4), which outperforms the state-of-the-art approach (Eyeriss) by 52.2% and 52.6%, respectively." @default.
- W2798317693 created "2018-05-07" @default.
- W2798317693 creator A5015435942 @default.
- W2798317693 creator A5018329525 @default.
- W2798317693 creator A5021570473 @default.
- W2798317693 creator A5034121567 @default.
- W2798317693 creator A5062689554 @default.
- W2798317693 creator A5069007431 @default.
- W2798317693 creator A5077006568 @default.
- W2798317693 date "2018-03-01" @default.
- W2798317693 modified "2023-10-05" @default.
- W2798317693 title "SmartShuttle: Optimizing off-chip memory accesses for deep learning accelerators" @default.
- W2798317693 cites W2094756095 @default.
- W2798317693 cites W2152839228 @default.
- W2798317693 cites W2155893237 @default.
- W2798317693 cites W2276486856 @default.
- W2798317693 cites W2285660444 @default.
- W2798317693 cites W2289252105 @default.
- W2798317693 cites W2612076670 @default.
- W2798317693 cites W2625457103 @default.
- W2798317693 cites W4240168186 @default.
- W2798317693 cites W4249932213 @default.
- W2798317693 doi "https://doi.org/10.23919/date.2018.8342033" @default.
- W2798317693 hasPublicationYear "2018" @default.
- W2798317693 type Work @default.
- W2798317693 sameAs 2798317693 @default.
- W2798317693 citedByCount "66" @default.
- W2798317693 countsByYear W27983176932018 @default.
- W2798317693 countsByYear W27983176932019 @default.
- W2798317693 countsByYear W27983176932020 @default.
- W2798317693 countsByYear W27983176932021 @default.
- W2798317693 countsByYear W27983176932022 @default.
- W2798317693 countsByYear W27983176932023 @default.
- W2798317693 crossrefType "proceedings-article" @default.
- W2798317693 hasAuthorship W2798317693A5015435942 @default.
- W2798317693 hasAuthorship W2798317693A5018329525 @default.
- W2798317693 hasAuthorship W2798317693A5021570473 @default.
- W2798317693 hasAuthorship W2798317693A5034121567 @default.
- W2798317693 hasAuthorship W2798317693A5062689554 @default.
- W2798317693 hasAuthorship W2798317693A5069007431 @default.
- W2798317693 hasAuthorship W2798317693A5077006568 @default.
- W2798317693 hasConcept C108583219 @default.
- W2798317693 hasConcept C113775141 @default.
- W2798317693 hasConcept C119599485 @default.
- W2798317693 hasConcept C127413603 @default.
- W2798317693 hasConcept C149635348 @default.
- W2798317693 hasConcept C154945302 @default.
- W2798317693 hasConcept C165005293 @default.
- W2798317693 hasConcept C173608175 @default.
- W2798317693 hasConcept C2742236 @default.
- W2798317693 hasConcept C41008148 @default.
- W2798317693 hasConcept C7366592 @default.
- W2798317693 hasConcept C76155785 @default.
- W2798317693 hasConcept C81363708 @default.
- W2798317693 hasConcept C9390403 @default.
- W2798317693 hasConceptScore W2798317693C108583219 @default.
- W2798317693 hasConceptScore W2798317693C113775141 @default.
- W2798317693 hasConceptScore W2798317693C119599485 @default.
- W2798317693 hasConceptScore W2798317693C127413603 @default.
- W2798317693 hasConceptScore W2798317693C149635348 @default.
- W2798317693 hasConceptScore W2798317693C154945302 @default.
- W2798317693 hasConceptScore W2798317693C165005293 @default.
- W2798317693 hasConceptScore W2798317693C173608175 @default.
- W2798317693 hasConceptScore W2798317693C2742236 @default.
- W2798317693 hasConceptScore W2798317693C41008148 @default.
- W2798317693 hasConceptScore W2798317693C7366592 @default.
- W2798317693 hasConceptScore W2798317693C76155785 @default.
- W2798317693 hasConceptScore W2798317693C81363708 @default.
- W2798317693 hasConceptScore W2798317693C9390403 @default.
- W2798317693 hasLocation W27983176931 @default.
- W2798317693 hasOpenAccess W2798317693 @default.
- W2798317693 hasPrimaryLocation W27983176931 @default.
- W2798317693 hasRelatedWork W1764185321 @default.
- W2798317693 hasRelatedWork W2037965222 @default.
- W2798317693 hasRelatedWork W2140848062 @default.
- W2798317693 hasRelatedWork W2354844719 @default.
- W2798317693 hasRelatedWork W2362222286 @default.
- W2798317693 hasRelatedWork W2374444867 @default.
- W2798317693 hasRelatedWork W2389537118 @default.
- W2798317693 hasRelatedWork W3094920005 @default.
- W2798317693 hasRelatedWork W3149602180 @default.
- W2798317693 hasRelatedWork W4210382982 @default.
- W2798317693 isParatext "false" @default.
- W2798317693 isRetracted "false" @default.
- W2798317693 magId "2798317693" @default.
- W2798317693 workType "article" @default.