Matches in SemOpenAlex for { <https://semopenalex.org/work/W2979411997> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2979411997 abstract "We present the design and optimization of a linear solver on General Purpose GPUs for the efficient and high-throughput evaluation of the marginalized graph kernel between pairs of labeled graphs. The solver implements a preconditioned conjugate gradient (PCG) method to compute the solution to a generalized Laplacian equation associated with the tensor product of two graphs. To cope with the gap between the instruction throughput and the memory bandwidth of current generation GPUs, our solver forms the tensor product linear system on-the-fly without storing it in memory when performing matrix-vector dot product operations in PCG. Such on-the-fly computation is accomplished by using threads in a warp to cooperatively stream the adjacency and edge label matrices of individual graphs by small square matrix blocks called tiles, which are then staged in registers and the shared memory for later reuse. Warps across a thread block can further share tiles via the shared memory to increase data reuse. We exploit the sparsity of the graphs hierarchically by storing only non-empty tiles using a coordinate format and nonzero elements within each tile using bitmaps. Besides, we propose a new partition-based reordering algorithm for aggregating nonzero elements of the graphs into fewer but denser tiles to improve the efficiency of the sparse format. We carry out extensive theoretical analyses on the graph tensor product primitives for tiles of various density and evaluate their performance on synthetic and real-world datasets. Our solver delivers three to four orders of magnitude speedup over existing CPU-based solvers such as GraKeL and GraphKernels. The capability of the solver enables kernel-based learning tasks at unprecedented scales." @default.
- W2979411997 created "2019-10-18" @default.
- W2979411997 creator A5019066795 @default.
- W2979411997 creator A5035719914 @default.
- W2979411997 creator A5040471064 @default.
- W2979411997 creator A5075496842 @default.
- W2979411997 date "2020-05-01" @default.
- W2979411997 modified "2023-09-23" @default.
- W2979411997 title "A High-Throughput Solver for Marginalized Graph Kernels on GPU" @default.
- W2979411997 cites W125936143 @default.
- W2979411997 cites W1973202568 @default.
- W2979411997 cites W1987427622 @default.
- W2979411997 cites W2002555321 @default.
- W2979411997 cites W2047259596 @default.
- W2979411997 cites W2056124433 @default.
- W2979411997 cites W2099438806 @default.
- W2979411997 cites W2104266030 @default.
- W2979411997 cites W2112882545 @default.
- W2979411997 cites W2114977680 @default.
- W2979411997 cites W2130479394 @default.
- W2979411997 cites W2414494135 @default.
- W2979411997 cites W2732233446 @default.
- W2979411997 cites W2760713876 @default.
- W2979411997 cites W2767891136 @default.
- W2979411997 cites W2897970024 @default.
- W2979411997 doi "https://doi.org/10.1109/ipdps47924.2020.00080" @default.
- W2979411997 hasPublicationYear "2020" @default.
- W2979411997 type Work @default.
- W2979411997 sameAs 2979411997 @default.
- W2979411997 citedByCount "7" @default.
- W2979411997 countsByYear W29794119972020 @default.
- W2979411997 countsByYear W29794119972021 @default.
- W2979411997 countsByYear W29794119972023 @default.
- W2979411997 crossrefType "proceedings-article" @default.
- W2979411997 hasAuthorship W2979411997A5019066795 @default.
- W2979411997 hasAuthorship W2979411997A5035719914 @default.
- W2979411997 hasAuthorship W2979411997A5040471064 @default.
- W2979411997 hasAuthorship W2979411997A5075496842 @default.
- W2979411997 hasBestOaLocation W29794119972 @default.
- W2979411997 hasConcept C121332964 @default.
- W2979411997 hasConcept C163716315 @default.
- W2979411997 hasConcept C173608175 @default.
- W2979411997 hasConcept C199360897 @default.
- W2979411997 hasConcept C2778770139 @default.
- W2979411997 hasConcept C41008148 @default.
- W2979411997 hasConcept C56372850 @default.
- W2979411997 hasConcept C62520636 @default.
- W2979411997 hasConcept C68339613 @default.
- W2979411997 hasConcept C80444323 @default.
- W2979411997 hasConceptScore W2979411997C121332964 @default.
- W2979411997 hasConceptScore W2979411997C163716315 @default.
- W2979411997 hasConceptScore W2979411997C173608175 @default.
- W2979411997 hasConceptScore W2979411997C199360897 @default.
- W2979411997 hasConceptScore W2979411997C2778770139 @default.
- W2979411997 hasConceptScore W2979411997C41008148 @default.
- W2979411997 hasConceptScore W2979411997C56372850 @default.
- W2979411997 hasConceptScore W2979411997C62520636 @default.
- W2979411997 hasConceptScore W2979411997C68339613 @default.
- W2979411997 hasConceptScore W2979411997C80444323 @default.
- W2979411997 hasLocation W29794119971 @default.
- W2979411997 hasLocation W29794119972 @default.
- W2979411997 hasLocation W29794119973 @default.
- W2979411997 hasLocation W29794119974 @default.
- W2979411997 hasLocation W29794119975 @default.
- W2979411997 hasOpenAccess W2979411997 @default.
- W2979411997 hasPrimaryLocation W29794119971 @default.
- W2979411997 hasRelatedWork W1509211761 @default.
- W2979411997 hasRelatedWork W1523072846 @default.
- W2979411997 hasRelatedWork W1969033192 @default.
- W2979411997 hasRelatedWork W2133693067 @default.
- W2979411997 hasRelatedWork W2142514638 @default.
- W2979411997 hasRelatedWork W2391299576 @default.
- W2979411997 hasRelatedWork W2526107130 @default.
- W2979411997 hasRelatedWork W2528505306 @default.
- W2979411997 hasRelatedWork W2612209529 @default.
- W2979411997 hasRelatedWork W2730282969 @default.
- W2979411997 isParatext "false" @default.
- W2979411997 isRetracted "false" @default.
- W2979411997 magId "2979411997" @default.
- W2979411997 workType "article" @default.