Matches in SemOpenAlex for { <https://semopenalex.org/work/W2806428272> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W2806428272 startingPage "3" @default.
- W2806428272 abstract "The convolution computation is widely used in many fields, especially in CNNs. Because of the rapid growth of the training data in CNNs, GPUs have been used for their acceleration and memory-efficient algorithms have been the focus of attention due to their high performance. In this paper, we propose two convolution kernels for single-channel and multi-channel convolution respectively. Our two methods achieve high performance by hiding the access delay of the global memory efficiently, and achieving high ratio of floating point Fused Multiply-Add operations per fetched data from the global memory. In comparison to the latest Cudnn library developed by Nvidia aimed to accelerate the deep-learning computation, the average performance improvement of our research is 2.6X for the single-channel, and 1.4X for the multi-channel." @default.
- W2806428272 created "2018-06-13" @default.
- W2806428272 creator A5011965834 @default.
- W2806428272 creator A5031341888 @default.
- W2806428272 creator A5041755728 @default.
- W2806428272 date "2018-04-15" @default.
- W2806428272 modified "2023-09-24" @default.
- W2806428272 title "Fast convolution kernels on pascal GPU with high memory efficiency" @default.
- W2806428272 cites W1667652561 @default.
- W2806428272 cites W1686810756 @default.
- W2806428272 cites W1922123711 @default.
- W2806428272 cites W2090593986 @default.
- W2806428272 cites W2097117768 @default.
- W2806428272 cites W2155893237 @default.
- W2806428272 cites W2163687928 @default.
- W2806428272 cites W2172654076 @default.
- W2806428272 cites W2178700889 @default.
- W2806428272 cites W2194775991 @default.
- W2806428272 cites W2481930807 @default.
- W2806428272 cites W2515863432 @default.
- W2806428272 cites W2617031569 @default.
- W2806428272 cites W2618530766 @default.
- W2806428272 cites W2963083845 @default.
- W2806428272 doi "https://doi.org/10.5555/3213069.3213072" @default.
- W2806428272 hasPublicationYear "2018" @default.
- W2806428272 type Work @default.
- W2806428272 sameAs 2806428272 @default.
- W2806428272 citedByCount "0" @default.
- W2806428272 crossrefType "proceedings-article" @default.
- W2806428272 hasAuthorship W2806428272A5011965834 @default.
- W2806428272 hasAuthorship W2806428272A5031341888 @default.
- W2806428272 hasAuthorship W2806428272A5041755728 @default.
- W2806428272 hasConcept C11413529 @default.
- W2806428272 hasConcept C114614502 @default.
- W2806428272 hasConcept C117896860 @default.
- W2806428272 hasConcept C120665830 @default.
- W2806428272 hasConcept C121332964 @default.
- W2806428272 hasConcept C127162648 @default.
- W2806428272 hasConcept C154945302 @default.
- W2806428272 hasConcept C173608175 @default.
- W2806428272 hasConcept C192209626 @default.
- W2806428272 hasConcept C199360897 @default.
- W2806428272 hasConcept C33923547 @default.
- W2806428272 hasConcept C41008148 @default.
- W2806428272 hasConcept C45347329 @default.
- W2806428272 hasConcept C45374587 @default.
- W2806428272 hasConcept C459310 @default.
- W2806428272 hasConcept C50644808 @default.
- W2806428272 hasConcept C74193536 @default.
- W2806428272 hasConcept C74650414 @default.
- W2806428272 hasConcept C75608658 @default.
- W2806428272 hasConcept C76155785 @default.
- W2806428272 hasConceptScore W2806428272C11413529 @default.
- W2806428272 hasConceptScore W2806428272C114614502 @default.
- W2806428272 hasConceptScore W2806428272C117896860 @default.
- W2806428272 hasConceptScore W2806428272C120665830 @default.
- W2806428272 hasConceptScore W2806428272C121332964 @default.
- W2806428272 hasConceptScore W2806428272C127162648 @default.
- W2806428272 hasConceptScore W2806428272C154945302 @default.
- W2806428272 hasConceptScore W2806428272C173608175 @default.
- W2806428272 hasConceptScore W2806428272C192209626 @default.
- W2806428272 hasConceptScore W2806428272C199360897 @default.
- W2806428272 hasConceptScore W2806428272C33923547 @default.
- W2806428272 hasConceptScore W2806428272C41008148 @default.
- W2806428272 hasConceptScore W2806428272C45347329 @default.
- W2806428272 hasConceptScore W2806428272C45374587 @default.
- W2806428272 hasConceptScore W2806428272C459310 @default.
- W2806428272 hasConceptScore W2806428272C50644808 @default.
- W2806428272 hasConceptScore W2806428272C74193536 @default.
- W2806428272 hasConceptScore W2806428272C74650414 @default.
- W2806428272 hasConceptScore W2806428272C75608658 @default.
- W2806428272 hasConceptScore W2806428272C76155785 @default.
- W2806428272 hasLocation W28064282721 @default.
- W2806428272 hasOpenAccess W2806428272 @default.
- W2806428272 hasPrimaryLocation W28064282721 @default.
- W2806428272 hasRelatedWork W199549905 @default.
- W2806428272 hasRelatedWork W2072912074 @default.
- W2806428272 hasRelatedWork W2129623330 @default.
- W2806428272 hasRelatedWork W2177655891 @default.
- W2806428272 hasRelatedWork W2178700889 @default.
- W2806428272 hasRelatedWork W2317796602 @default.
- W2806428272 hasRelatedWork W2605739168 @default.
- W2806428272 hasRelatedWork W2617031569 @default.
- W2806428272 hasRelatedWork W2760845253 @default.
- W2806428272 hasRelatedWork W2800485305 @default.
- W2806428272 hasRelatedWork W2886776382 @default.
- W2806428272 hasRelatedWork W2945580137 @default.
- W2806428272 hasRelatedWork W2972202684 @default.
- W2806428272 hasRelatedWork W2999034465 @default.
- W2806428272 hasRelatedWork W3011884329 @default.
- W2806428272 hasRelatedWork W3040913567 @default.
- W2806428272 hasRelatedWork W3090303800 @default.
- W2806428272 hasRelatedWork W3095498018 @default.
- W2806428272 hasRelatedWork W3157543420 @default.
- W2806428272 hasRelatedWork W3166036439 @default.
- W2806428272 isParatext "false" @default.
- W2806428272 isRetracted "false" @default.
- W2806428272 magId "2806428272" @default.
- W2806428272 workType "article" @default.