FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2094756095> ?p ?o ?g. }

• W2094756095 abstract "Convolutional neural network (CNN) has been widely employed for image recognition because it can achieve high accuracy by emulating behavior of optic nerves in living creatures. Recently, rapid growth of modern applications based on deep learning algorithms has further improved research and implementations. Especially, various accelerators for deep CNN have been proposed based on FPGA platform because it has advantages of high performance, reconfigurability, and fast development round, etc. Although current FPGA accelerators have demonstrated better performance over generic processors, the accelerator design space has not been well exploited. One critical problem is that the computation throughput may not well match the memory bandwidth provided an FPGA platform. Consequently, existing approaches cannot achieve best performance due to under-utilization of either logic resource or memory bandwidth. At the same time, the increasing complexity and scalability of deep learning applications aggravate this problem. In order to overcome this problem, we propose an analytical design scheme using the roofline model. For any solution of a CNN design, we quantitatively analyze its computing throughput and required memory bandwidth using various optimization techniques, such as loop tiling and transformation. Then, with the help of rooine model, we can identify the solution with best performance and lowest FPGA resource requirement. As a case study, we implement a CNN accelerator on a VC707 FPGA board and compare it to previous approaches. Our implementation achieves a peak performance of 61.62 GFLOPS under 100MHz working frequency, which outperform previous approaches significantly." @default.
• W2094756095 created "2016-06-24" @default.
• W2094756095 creator A5009018872 @default.
• W2094756095 creator A5016776689 @default.
• W2094756095 creator A5047788295 @default.
• W2094756095 creator A5050841063 @default.
• W2094756095 creator A5057276000 @default.
• W2094756095 creator A5066056671 @default.
• W2094756095 date "2015-02-22" @default.
• W2094756095 modified "2023-10-16" @default.
• W2094756095 title "Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks" @default.
• W2094756095 cites W1005811612 @default.
• W2094756095 cites W1983364832 @default.
• W2094756095 cites W1990315422 @default.
• W2094756095 cites W1994197834 @default.
• W2094756095 cites W2002555321 @default.
• W2094756095 cites W2053968820 @default.
• W2094756095 cites W2096645269 @default.
• W2094756095 cites W2112796928 @default.
• W2094756095 cites W2115572397 @default.
• W2094756095 cites W2117696986 @default.
• W2094756095 cites W2141280299 @default.
• W2094756095 cites W3004171485 @default.
• W2094756095 cites W4212788319 @default.
• W2094756095 doi "https://doi.org/10.1145/2684746.2689060" @default.
• W2094756095 hasPublicationYear "2015" @default.
• W2094756095 type Work @default.
• W2094756095 sameAs 2094756095 @default.
• W2094756095 citedByCount "1430" @default.
• W2094756095 countsByYear W20947560952012 @default.
• W2094756095 countsByYear W20947560952015 @default.
• W2094756095 countsByYear W20947560952016 @default.
• W2094756095 countsByYear W20947560952017 @default.
• W2094756095 countsByYear W20947560952018 @default.
• W2094756095 countsByYear W20947560952019 @default.
• W2094756095 countsByYear W20947560952020 @default.
• W2094756095 countsByYear W20947560952021 @default.
• W2094756095 countsByYear W20947560952022 @default.
• W2094756095 countsByYear W20947560952023 @default.
• W2094756095 crossrefType "proceedings-article" @default.
• W2094756095 hasAuthorship W2094756095A5009018872 @default.
• W2094756095 hasAuthorship W2094756095A5016776689 @default.
• W2094756095 hasAuthorship W2094756095A5047788295 @default.
• W2094756095 hasAuthorship W2094756095A5050841063 @default.
• W2094756095 hasAuthorship W2094756095A5057276000 @default.
• W2094756095 hasAuthorship W2094756095A5066056671 @default.
• W2094756095 hasConcept C108583219 @default.
• W2094756095 hasConcept C113775141 @default.
• W2094756095 hasConcept C118524514 @default.
• W2094756095 hasConcept C149635348 @default.
• W2094756095 hasConcept C154945302 @default.
• W2094756095 hasConcept C157764524 @default.
• W2094756095 hasConcept C188045654 @default.
• W2094756095 hasConcept C2776221188 @default.
• W2094756095 hasConcept C2776257435 @default.
• W2094756095 hasConcept C2780149590 @default.
• W2094756095 hasConcept C31258907 @default.
• W2094756095 hasConcept C41008148 @default.
• W2094756095 hasConcept C42935608 @default.
• W2094756095 hasConcept C48044578 @default.
• W2094756095 hasConcept C555944384 @default.
• W2094756095 hasConcept C76155785 @default.
• W2094756095 hasConcept C77088390 @default.
• W2094756095 hasConcept C81363708 @default.
• W2094756095 hasConcept C9390403 @default.
• W2094756095 hasConceptScore W2094756095C108583219 @default.
• W2094756095 hasConceptScore W2094756095C113775141 @default.
• W2094756095 hasConceptScore W2094756095C118524514 @default.
• W2094756095 hasConceptScore W2094756095C149635348 @default.
• W2094756095 hasConceptScore W2094756095C154945302 @default.
• W2094756095 hasConceptScore W2094756095C157764524 @default.
• W2094756095 hasConceptScore W2094756095C188045654 @default.
• W2094756095 hasConceptScore W2094756095C2776221188 @default.
• W2094756095 hasConceptScore W2094756095C2776257435 @default.
• W2094756095 hasConceptScore W2094756095C2780149590 @default.
• W2094756095 hasConceptScore W2094756095C31258907 @default.
• W2094756095 hasConceptScore W2094756095C41008148 @default.
• W2094756095 hasConceptScore W2094756095C42935608 @default.
• W2094756095 hasConceptScore W2094756095C48044578 @default.
• W2094756095 hasConceptScore W2094756095C555944384 @default.
• W2094756095 hasConceptScore W2094756095C76155785 @default.
• W2094756095 hasConceptScore W2094756095C77088390 @default.
• W2094756095 hasConceptScore W2094756095C81363708 @default.
• W2094756095 hasConceptScore W2094756095C9390403 @default.
• W2094756095 hasLocation W20947560951 @default.
• W2094756095 hasOpenAccess W2094756095 @default.
• W2094756095 hasPrimaryLocation W20947560951 @default.
• W2094756095 hasRelatedWork W2134422574 @default.
• W2094756095 hasRelatedWork W2295680811 @default.
• W2094756095 hasRelatedWork W2320205417 @default.
• W2094756095 hasRelatedWork W2524802307 @default.
• W2094756095 hasRelatedWork W2980171969 @default.
• W2094756095 hasRelatedWork W3003304268 @default.
• W2094756095 hasRelatedWork W4214822271 @default.
• W2094756095 hasRelatedWork W4285708690 @default.
• W2094756095 hasRelatedWork W4319662858 @default.
• W2094756095 hasRelatedWork W4361251788 @default.
• W2094756095 isParatext "false" @default.
• W2094756095 isRetracted "false" @default.
• W2094756095 magId "2094756095" @default.

• next