Matches in SemOpenAlex for { <https://semopenalex.org/work/W2954905927> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W2954905927 abstract "Graph Convolutional Networks (GCNs) are recently getting much attention in bioinformatics and chemoinformatics as a state-of-the-art machine learning approach with high accuracy. GCNs process convolutional operations along with graph structures, and GPUs are used to process enormous operations including sparse-dense matrix multiplication (SpMM) when the graph structure is expressed as an adjacency matrix with sparse matrix format. However, the SpMM operation on small graph, where the number of nodes is tens or hundreds, hardly exploits high parallelism or compute power of GPU. Therefore, SpMM becomes a bottleneck of training and inference in GCNs applications. In order to improve the performance of GCNs applications, we propose new SpMM algorithm especially for small sparse matrix and Batched SpMM, which exploits high parallelism of GPU by processing multiple SpMM operations with single CUDA kernel. To the best of our knowledge, this is the first work of batched approach for SpMM. We evaluated the performance of the GCNs application on TSUBAME3.0 implementing NVIDIA Tesla P100 GPU, and our batched approach shows significant speedups of up to 1.59x and 1.37x in training and inference, respectively." @default.
- W2954905927 created "2019-07-12" @default.
- W2954905927 creator A5006972521 @default.
- W2954905927 creator A5013489463 @default.
- W2954905927 creator A5017240065 @default.
- W2954905927 creator A5069048439 @default.
- W2954905927 date "2019-03-27" @default.
- W2954905927 modified "2023-09-27" @default.
- W2954905927 title "Batched Sparse Matrix Multiplication for Accelerating Graph Convolutional Networks" @default.
- W2954905927 cites W1028753842 @default.
- W2954905927 cites W1545561241 @default.
- W2954905927 cites W2006682733 @default.
- W2954905927 cites W2076435198 @default.
- W2954905927 cites W2099021415 @default.
- W2954905927 cites W2134237243 @default.
- W2954905927 cites W2138215414 @default.
- W2954905927 cites W2244807774 @default.
- W2954905927 cites W2499931820 @default.
- W2954905927 cites W2519887557 @default.
- W2954905927 cites W2531348778 @default.
- W2954905927 cites W2559809597 @default.
- W2954905927 cites W2566719947 @default.
- W2954905927 cites W2601355684 @default.
- W2954905927 cites W2604314403 @default.
- W2954905927 cites W2736056409 @default.
- W2954905927 cites W2753962198 @default.
- W2954905927 cites W2765161574 @default.
- W2954905927 cites W2808128431 @default.
- W2954905927 cites W2952254971 @default.
- W2954905927 cites W2964113829 @default.
- W2954905927 cites W2964337156 @default.
- W2954905927 doi "https://doi.org/10.48550/arxiv.1903.11409" @default.
- W2954905927 hasPublicationYear "2019" @default.
- W2954905927 type Work @default.
- W2954905927 sameAs 2954905927 @default.
- W2954905927 citedByCount "0" @default.
- W2954905927 crossrefType "posted-content" @default.
- W2954905927 hasAuthorship W2954905927A5006972521 @default.
- W2954905927 hasAuthorship W2954905927A5013489463 @default.
- W2954905927 hasAuthorship W2954905927A5017240065 @default.
- W2954905927 hasAuthorship W2954905927A5069048439 @default.
- W2954905927 hasBestOaLocation W29549059271 @default.
- W2954905927 hasConcept C11413529 @default.
- W2954905927 hasConcept C114614502 @default.
- W2954905927 hasConcept C121332964 @default.
- W2954905927 hasConcept C121684516 @default.
- W2954905927 hasConcept C132525143 @default.
- W2954905927 hasConcept C154945302 @default.
- W2954905927 hasConcept C163716315 @default.
- W2954905927 hasConcept C17349429 @default.
- W2954905927 hasConcept C173608175 @default.
- W2954905927 hasConcept C180356752 @default.
- W2954905927 hasConcept C21442007 @default.
- W2954905927 hasConcept C2776214188 @default.
- W2954905927 hasConcept C33923547 @default.
- W2954905927 hasConcept C41008148 @default.
- W2954905927 hasConcept C50630238 @default.
- W2954905927 hasConcept C56372850 @default.
- W2954905927 hasConcept C62520636 @default.
- W2954905927 hasConcept C74193536 @default.
- W2954905927 hasConcept C80444323 @default.
- W2954905927 hasConcept C84114770 @default.
- W2954905927 hasConceptScore W2954905927C11413529 @default.
- W2954905927 hasConceptScore W2954905927C114614502 @default.
- W2954905927 hasConceptScore W2954905927C121332964 @default.
- W2954905927 hasConceptScore W2954905927C121684516 @default.
- W2954905927 hasConceptScore W2954905927C132525143 @default.
- W2954905927 hasConceptScore W2954905927C154945302 @default.
- W2954905927 hasConceptScore W2954905927C163716315 @default.
- W2954905927 hasConceptScore W2954905927C17349429 @default.
- W2954905927 hasConceptScore W2954905927C173608175 @default.
- W2954905927 hasConceptScore W2954905927C180356752 @default.
- W2954905927 hasConceptScore W2954905927C21442007 @default.
- W2954905927 hasConceptScore W2954905927C2776214188 @default.
- W2954905927 hasConceptScore W2954905927C33923547 @default.
- W2954905927 hasConceptScore W2954905927C41008148 @default.
- W2954905927 hasConceptScore W2954905927C50630238 @default.
- W2954905927 hasConceptScore W2954905927C56372850 @default.
- W2954905927 hasConceptScore W2954905927C62520636 @default.
- W2954905927 hasConceptScore W2954905927C74193536 @default.
- W2954905927 hasConceptScore W2954905927C80444323 @default.
- W2954905927 hasConceptScore W2954905927C84114770 @default.
- W2954905927 hasLocation W29549059271 @default.
- W2954905927 hasOpenAccess W2954905927 @default.
- W2954905927 hasPrimaryLocation W29549059271 @default.
- W2954905927 hasRelatedWork W2011159963 @default.
- W2954905927 hasRelatedWork W2013693725 @default.
- W2954905927 hasRelatedWork W2153645222 @default.
- W2954905927 hasRelatedWork W2985041356 @default.
- W2954905927 hasRelatedWork W3015799701 @default.
- W2954905927 hasRelatedWork W3099744258 @default.
- W2954905927 hasRelatedWork W3176486103 @default.
- W2954905927 hasRelatedWork W4200089575 @default.
- W2954905927 hasRelatedWork W4250047567 @default.
- W2954905927 hasRelatedWork W2127872620 @default.
- W2954905927 isParatext "false" @default.
- W2954905927 isRetracted "false" @default.
- W2954905927 magId "2954905927" @default.
- W2954905927 workType "article" @default.