Matches in SemOpenAlex for { <https://semopenalex.org/work/W4294032011> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W4294032011 abstract "Transposed convolution has been prevailing in convolutional neural networks (CNNs), playing an important role in multiple scenarios such as image segmentation and back-propagation process of training CNNs. This mainly benefits from the ability to up-sample the input feature maps by interpolating new information from the input feature pixels. However, the backward-stencil computation constrains its performance and hindered its wide application in diverse platforms. Moreover, in contrast to the efforts on accelerating the convolution, there is a rare investigation on the acceleration of transposed convolution that is identically compute-intensive as the former. For acceleration of transposed convolution, we propose an intermediate-centric dataflow scheme, in which we decouple the generation of the intermediate patch from its further process, aiming to efficiently perform the backward-stencil computation . The intermediate-centric dataflow breaks the transposed convolution into several phases/stages, achieving feeding the input feature maps and performing the backward-stencil computation in a pipelining manner. It also provides four-degree computation parallelism and efficient data reuse of input feature maps/weights. Furthermore, we also theoretically analyze the irregular data dependence leveraging the polyhedral model, which constrains the parallel computing of transposed convolution. Additionally, we devise an optimization problem to explore the design space and automatically generate the optimal design configurations for different transposed convolutional layers and hardware platforms. By selecting the representative transposed convolutional layers from DCGAN, FSRCNN, and FCN, we generate the corresponding accelerator arrays of intermediate-centric dataflow on the Xilinx Alveo U200 platform and reach the performance of 3.92 TOPS, 2.72 TOPS, and 4.76 TOPS, respectively." @default.
- W4294032011 created "2022-09-01" @default.
- W4294032011 creator A5023468643 @default.
- W4294032011 creator A5031612856 @default.
- W4294032011 creator A5049858651 @default.
- W4294032011 creator A5054412582 @default.
- W4294032011 date "2022-09-01" @default.
- W4294032011 modified "2023-09-27" @default.
- W4294032011 title "An <i>Intermediate-Centric</i> Dataflow for Transposed Convolution Acceleration on FPGA" @default.
- W4294032011 cites W1903029394 @default.
- W4294032011 cites W1983394510 @default.
- W4294032011 cites W2047920195 @default.
- W4294032011 cites W2076517649 @default.
- W4294032011 cites W2094756095 @default.
- W4294032011 cites W2276486856 @default.
- W4294032011 cites W2331128040 @default.
- W4294032011 cites W2503339013 @default.
- W4294032011 cites W2606722458 @default.
- W4294032011 cites W2625954420 @default.
- W4294032011 cites W2727238169 @default.
- W4294032011 cites W2891976760 @default.
- W4294032011 cites W2895738878 @default.
- W4294032011 cites W2899824860 @default.
- W4294032011 cites W2899915146 @default.
- W4294032011 cites W2906403229 @default.
- W4294032011 cites W2942756726 @default.
- W4294032011 cites W2963577671 @default.
- W4294032011 cites W2964231161 @default.
- W4294032011 cites W3040533934 @default.
- W4294032011 cites W3130920634 @default.
- W4294032011 cites W3137762252 @default.
- W4294032011 doi "https://doi.org/10.1145/3561053" @default.
- W4294032011 hasPublicationYear "2022" @default.
- W4294032011 type Work @default.
- W4294032011 citedByCount "0" @default.
- W4294032011 crossrefType "journal-article" @default.
- W4294032011 hasAuthorship W4294032011A5023468643 @default.
- W4294032011 hasAuthorship W4294032011A5031612856 @default.
- W4294032011 hasAuthorship W4294032011A5049858651 @default.
- W4294032011 hasAuthorship W4294032011A5054412582 @default.
- W4294032011 hasBestOaLocation W42940320111 @default.
- W4294032011 hasConcept C11413529 @default.
- W4294032011 hasConcept C114614502 @default.
- W4294032011 hasConcept C117896860 @default.
- W4294032011 hasConcept C121332964 @default.
- W4294032011 hasConcept C138885662 @default.
- W4294032011 hasConcept C154945302 @default.
- W4294032011 hasConcept C173608175 @default.
- W4294032011 hasConcept C2776401178 @default.
- W4294032011 hasConcept C33923547 @default.
- W4294032011 hasConcept C41008148 @default.
- W4294032011 hasConcept C41895202 @default.
- W4294032011 hasConcept C45347329 @default.
- W4294032011 hasConcept C45374587 @default.
- W4294032011 hasConcept C459310 @default.
- W4294032011 hasConcept C50644808 @default.
- W4294032011 hasConcept C68339613 @default.
- W4294032011 hasConcept C74193536 @default.
- W4294032011 hasConcept C74650414 @default.
- W4294032011 hasConcept C76752949 @default.
- W4294032011 hasConcept C81363708 @default.
- W4294032011 hasConcept C96324660 @default.
- W4294032011 hasConceptScore W4294032011C11413529 @default.
- W4294032011 hasConceptScore W4294032011C114614502 @default.
- W4294032011 hasConceptScore W4294032011C117896860 @default.
- W4294032011 hasConceptScore W4294032011C121332964 @default.
- W4294032011 hasConceptScore W4294032011C138885662 @default.
- W4294032011 hasConceptScore W4294032011C154945302 @default.
- W4294032011 hasConceptScore W4294032011C173608175 @default.
- W4294032011 hasConceptScore W4294032011C2776401178 @default.
- W4294032011 hasConceptScore W4294032011C33923547 @default.
- W4294032011 hasConceptScore W4294032011C41008148 @default.
- W4294032011 hasConceptScore W4294032011C41895202 @default.
- W4294032011 hasConceptScore W4294032011C45347329 @default.
- W4294032011 hasConceptScore W4294032011C45374587 @default.
- W4294032011 hasConceptScore W4294032011C459310 @default.
- W4294032011 hasConceptScore W4294032011C50644808 @default.
- W4294032011 hasConceptScore W4294032011C68339613 @default.
- W4294032011 hasConceptScore W4294032011C74193536 @default.
- W4294032011 hasConceptScore W4294032011C74650414 @default.
- W4294032011 hasConceptScore W4294032011C76752949 @default.
- W4294032011 hasConceptScore W4294032011C81363708 @default.
- W4294032011 hasConceptScore W4294032011C96324660 @default.
- W4294032011 hasLocation W42940320111 @default.
- W4294032011 hasOpenAccess W4294032011 @default.
- W4294032011 hasPrimaryLocation W42940320111 @default.
- W4294032011 hasRelatedWork W1572523360 @default.
- W4294032011 hasRelatedWork W2068258741 @default.
- W4294032011 hasRelatedWork W2537455120 @default.
- W4294032011 hasRelatedWork W2767899175 @default.
- W4294032011 hasRelatedWork W2796257099 @default.
- W4294032011 hasRelatedWork W2804132206 @default.
- W4294032011 hasRelatedWork W3122909934 @default.
- W4294032011 hasRelatedWork W4200123712 @default.
- W4294032011 hasRelatedWork W4294032011 @default.
- W4294032011 hasRelatedWork W4323831447 @default.
- W4294032011 isParatext "false" @default.
- W4294032011 isRetracted "false" @default.
- W4294032011 workType "article" @default.