FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2896983500> ?p ?o ?g. }

• W2896983500 endingPage "2085" @default.
• W2896983500 startingPage "2072" @default.
• W2896983500 abstract "With the recent advancement of multilayer convolutional neural networks (CNNs) and fully connected networks (FCNs), deep learning has achieved amazing success in many areas, especially in visual content understanding and classification. To improve the performance and energy efficiency of the computation-demanding CNN, the FPGA-based acceleration emerges as one of the most attractive alternatives. In this paper, we design and implement Caffeine, a hardware/software co-designed library to efficiently accelerate the entire CNN and FCN on FPGAs. First, we propose a uniformed convolutional matrix-multiplication representation for both computation-bound convolutional layers and communication-bound FCN layers. Based on this representation, we optimize the accelerator microarchitecture and maximize the underlying FPGA computing and bandwidth resource utilization based on a revised roofline model. Moreover, we design an automation flow to directly compile highlevel network definitions to the final FPGA accelerator. As a case study, we integrate Caffeine into the industry-standard software deep learning framework Caffe. We evaluate Caffeine and its integration with Caffe by implementing VGG16 and AlexNet networks on multiple FPGA platforms. Caffeine achieves a peak performance of 1460 giga fixed point operations per second on a medium-sized Xilinx KU060 FPGA board; to our knowledge, this is the best published result. It achieves more than 100× speedup on FCN layers over prior FPGA accelerators. An end-to-end evaluation with Caffe integration shows up to 29× and 150× performance and energy gains over Caffe on a 12-core Xeon server, and 5.7× better energy efficiency over the GPU implementation. Performance projections for a system with a high-end FPGA (Virtex7 690t) show even higher gains." @default.
• W2896983500 created "2018-10-26" @default.
• W2896983500 creator A5009018872 @default.
• W2896983500 creator A5016776689 @default.
• W2896983500 creator A5063866156 @default.
• W2896983500 creator A5065889904 @default.
• W2896983500 creator A5066056671 @default.
• W2896983500 creator A5069821335 @default.
• W2896983500 date "2019-11-01" @default.
• W2896983500 modified "2023-10-12" @default.
• W2896983500 title "Caffeine: Toward Uniformed Representation and Acceleration for Deep Convolutional Neural Networks" @default.
• W2896983500 cites W1677182931 @default.
• W2896983500 cites W1849277567 @default.
• W2896983500 cites W1970088388 @default.
• W2896983500 cites W1982052956 @default.
• W2896983500 cites W1983364832 @default.
• W2896983500 cites W1990315422 @default.
• W2896983500 cites W2002555321 @default.
• W2896983500 cites W2009832130 @default.
• W2896983500 cites W2012897754 @default.
• W2896983500 cites W2015048865 @default.
• W2896983500 cites W2053968820 @default.
• W2896983500 cites W2096645269 @default.
• W2896983500 cites W2097117768 @default.
• W2896983500 cites W2102605133 @default.
• W2896983500 cites W2115572397 @default.
• W2896983500 cites W2117696986 @default.
• W2896983500 cites W2145287260 @default.
• W2896983500 cites W2160815625 @default.
• W2896983500 cites W2170503676 @default.
• W2896983500 cites W2276486856 @default.
• W2896983500 cites W2294282016 @default.
• W2896983500 cites W2405102949 @default.
• W2896983500 cites W2520083297 @default.
• W2896983500 cites W2584311934 @default.
• W2896983500 cites W2584616277 @default.
• W2896983500 cites W2585560244 @default.
• W2896983500 cites W2585774018 @default.
• W2896983500 cites W2606722458 @default.
• W2896983500 cites W2625954420 @default.
• W2896983500 cites W4212788319 @default.
• W2896983500 doi "https://doi.org/10.1109/tcad.2017.2785257" @default.
• W2896983500 hasPublicationYear "2019" @default.
• W2896983500 type Work @default.
• W2896983500 sameAs 2896983500 @default.
• W2896983500 citedByCount "176" @default.
• W2896983500 countsByYear W28969835002019 @default.
• W2896983500 countsByYear W28969835002020 @default.
• W2896983500 countsByYear W28969835002021 @default.
• W2896983500 countsByYear W28969835002022 @default.
• W2896983500 countsByYear W28969835002023 @default.
• W2896983500 crossrefType "journal-article" @default.
• W2896983500 hasAuthorship W2896983500A5009018872 @default.
• W2896983500 hasAuthorship W2896983500A5016776689 @default.
• W2896983500 hasAuthorship W2896983500A5063866156 @default.
• W2896983500 hasAuthorship W2896983500A5065889904 @default.
• W2896983500 hasAuthorship W2896983500A5066056671 @default.
• W2896983500 hasAuthorship W2896983500A5069821335 @default.
• W2896983500 hasBestOaLocation W28969835001 @default.
• W2896983500 hasConcept C108583219 @default.
• W2896983500 hasConcept C118524514 @default.
• W2896983500 hasConcept C13164978 @default.
• W2896983500 hasConcept C145108525 @default.
• W2896983500 hasConcept C149635348 @default.
• W2896983500 hasConcept C154945302 @default.
• W2896983500 hasConcept C173608175 @default.
• W2896983500 hasConcept C41008148 @default.
• W2896983500 hasConcept C42935608 @default.
• W2896983500 hasConcept C50644808 @default.
• W2896983500 hasConcept C68339613 @default.
• W2896983500 hasConcept C81363708 @default.
• W2896983500 hasConceptScore W2896983500C108583219 @default.
• W2896983500 hasConceptScore W2896983500C118524514 @default.
• W2896983500 hasConceptScore W2896983500C13164978 @default.
• W2896983500 hasConceptScore W2896983500C145108525 @default.
• W2896983500 hasConceptScore W2896983500C149635348 @default.
• W2896983500 hasConceptScore W2896983500C154945302 @default.
• W2896983500 hasConceptScore W2896983500C173608175 @default.
• W2896983500 hasConceptScore W2896983500C41008148 @default.
• W2896983500 hasConceptScore W2896983500C42935608 @default.
• W2896983500 hasConceptScore W2896983500C50644808 @default.
• W2896983500 hasConceptScore W2896983500C68339613 @default.
• W2896983500 hasConceptScore W2896983500C81363708 @default.
• W2896983500 hasFunder F4320309469 @default.
• W2896983500 hasFunder F4320321001 @default.
• W2896983500 hasFunder F4320322725 @default.
• W2896983500 hasIssue "11" @default.
• W2896983500 hasLocation W28969835001 @default.
• W2896983500 hasOpenAccess W2896983500 @default.
• W2896983500 hasPrimaryLocation W28969835001 @default.
• W2896983500 hasRelatedWork W1608572506 @default.
• W2896983500 hasRelatedWork W1967938402 @default.
• W2896983500 hasRelatedWork W2160474882 @default.
• W2896983500 hasRelatedWork W2218038495 @default.
• W2896983500 hasRelatedWork W229781084 @default.
• W2896983500 hasRelatedWork W2386041993 @default.
• W2896983500 hasRelatedWork W2800704601 @default.
• W2896983500 hasRelatedWork W2943610686 @default.

• next