FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3022173611> ?p ?o ?g. }

• W3022173611 endingPage "3069" @default.
• W3022173611 startingPage "3056" @default.
• W3022173611 abstract "In this paper, we propose O <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>4</sup> -DNN, a high-performance FPGA-based architecture for convolutional neural network (CNN) accelerators relying on operation packing and out-of-order (OoO) execution for DSP blocks augmented with LUT-based glue logic. The high-level architecture is comprised of a systolic array of processing elements (PEs), supporting output stationary dataflow. In this architecture, the computational unit of each PE is realized by using a DSP block as well as a small number of LUTs. Given the limited number of DSP blocks in FPGAs, the combination (DSP block and some LUTs) provides more computational power obtainable through each DSP block. The proposed computational unit performs eight convolutional operations on five input operands where one of them is an 8-bit weight and the others are four 8-bit input feature (IF) maps. In addition, to improve the energy efficiency of the proposed computational unit, we present an approximate form of the unit suitable for neural network applications. To reduce the memory bandwidth as well as increase the utilization of the computational units, a data reusing technique based on the weight sharing is also presented. To improve the performance of the proposed computational unit further, an addressing approach for computing the partial sums out-of-order is proposed. The efficacy of the architecture is assessed using two FPGA devices executing four state-of-the-art neural networks. Experimental results show that this architecture leads to, on average (up to), 2.5× (3.44×) higher throughput compared to a baseline structure. In addition, on average (maximum of), 12% (40%) energy efficiency improvement is achievable by employing the O <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>4</sup> -DNN compared to the baseline structure." @default.
• W3022173611 created "2020-05-13" @default.
• W3022173611 creator A5042809860 @default.
• W3022173611 creator A5044650311 @default.
• W3022173611 creator A5067209238 @default.
• W3022173611 creator A5074063358 @default.
• W3022173611 date "2020-09-01" @default.
• W3022173611 modified "2023-09-30" @default.
• W3022173611 title "O⁴-DNN: A Hybrid DSP-LUT-Based Processing Unit With Operation Packing and Out-of-Order Execution for Efficient Realization of Convolutional Neural Networks on FPGA Devices" @default.
• W3022173611 cites W1934410531 @default.
• W3022173611 cites W2002555321 @default.
• W3022173611 cites W2117539524 @default.
• W3022173611 cites W2160815625 @default.
• W3022173611 cites W2194775991 @default.
• W3022173611 cites W2583383421 @default.
• W3022173611 cites W2604319603 @default.
• W3022173611 cites W2606722458 @default.
• W3022173611 cites W2625954420 @default.
• W3022173611 cites W2725615981 @default.
• W3022173611 cites W2762374354 @default.
• W3022173611 cites W2795915628 @default.
• W3022173611 cites W2795961274 @default.
• W3022173611 cites W2903688003 @default.
• W3022173611 cites W2903735800 @default.
• W3022173611 cites W2949275038 @default.
• W3022173611 cites W2951537853 @default.
• W3022173611 cites W2963594949 @default.
• W3022173611 cites W4230481654 @default.
• W3022173611 doi "https://doi.org/10.1109/tcsi.2020.2986350" @default.
• W3022173611 hasPublicationYear "2020" @default.
• W3022173611 type Work @default.
• W3022173611 sameAs 3022173611 @default.
• W3022173611 citedByCount "2" @default.
• W3022173611 countsByYear W30221736112021 @default.
• W3022173611 countsByYear W30221736112023 @default.
• W3022173611 crossrefType "journal-article" @default.
• W3022173611 hasAuthorship W3022173611A5042809860 @default.
• W3022173611 hasAuthorship W3022173611A5044650311 @default.
• W3022173611 hasAuthorship W3022173611A5067209238 @default.
• W3022173611 hasAuthorship W3022173611A5074063358 @default.
• W3022173611 hasBestOaLocation W30221736111 @default.
• W3022173611 hasConcept C113775141 @default.
• W3022173611 hasConcept C118524514 @default.
• W3022173611 hasConcept C134835016 @default.
• W3022173611 hasConcept C149635348 @default.
• W3022173611 hasConcept C154945302 @default.
• W3022173611 hasConcept C173608175 @default.
• W3022173611 hasConcept C199360897 @default.
• W3022173611 hasConcept C2524010 @default.
• W3022173611 hasConcept C2777210771 @default.
• W3022173611 hasConcept C33923547 @default.
• W3022173611 hasConcept C41008148 @default.
• W3022173611 hasConcept C42935608 @default.
• W3022173611 hasConcept C50644808 @default.
• W3022173611 hasConcept C55526617 @default.
• W3022173611 hasConcept C81363708 @default.
• W3022173611 hasConcept C84462506 @default.
• W3022173611 hasConcept C9390403 @default.
• W3022173611 hasConceptScore W3022173611C113775141 @default.
• W3022173611 hasConceptScore W3022173611C118524514 @default.
• W3022173611 hasConceptScore W3022173611C134835016 @default.
• W3022173611 hasConceptScore W3022173611C149635348 @default.
• W3022173611 hasConceptScore W3022173611C154945302 @default.
• W3022173611 hasConceptScore W3022173611C173608175 @default.
• W3022173611 hasConceptScore W3022173611C199360897 @default.
• W3022173611 hasConceptScore W3022173611C2524010 @default.
• W3022173611 hasConceptScore W3022173611C2777210771 @default.
• W3022173611 hasConceptScore W3022173611C33923547 @default.
• W3022173611 hasConceptScore W3022173611C41008148 @default.
• W3022173611 hasConceptScore W3022173611C42935608 @default.
• W3022173611 hasConceptScore W3022173611C50644808 @default.
• W3022173611 hasConceptScore W3022173611C55526617 @default.
• W3022173611 hasConceptScore W3022173611C81363708 @default.
• W3022173611 hasConceptScore W3022173611C84462506 @default.
• W3022173611 hasConceptScore W3022173611C9390403 @default.
• W3022173611 hasFunder F4320306076 @default.
• W3022173611 hasIssue "9" @default.
• W3022173611 hasLocation W30221736111 @default.
• W3022173611 hasOpenAccess W3022173611 @default.
• W3022173611 hasPrimaryLocation W30221736111 @default.
• W3022173611 hasRelatedWork W2040548413 @default.
• W3022173611 hasRelatedWork W2129894819 @default.
• W3022173611 hasRelatedWork W2143940934 @default.
• W3022173611 hasRelatedWork W2339728242 @default.
• W3022173611 hasRelatedWork W2350861609 @default.
• W3022173611 hasRelatedWork W2352017551 @default.
• W3022173611 hasRelatedWork W2363310833 @default.
• W3022173611 hasRelatedWork W2524802307 @default.
• W3022173611 hasRelatedWork W4213033874 @default.
• W3022173611 hasRelatedWork W4308216800 @default.
• W3022173611 hasVolume "67" @default.
• W3022173611 isParatext "false" @default.
• W3022173611 isRetracted "false" @default.
• W3022173611 magId "3022173611" @default.
• W3022173611 workType "article" @default.