Matches in SemOpenAlex for { <https://semopenalex.org/work/W2985111225> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2985111225 endingPage "107084" @default.
- W2985111225 startingPage "107084" @default.
- W2985111225 abstract "Recently, graph neural networks (GNNs) have shown to be effective in learning representative graph features. However, current pooling-based strategies for graph classification lack efficient utilization of graph representation information in which each node and layer have the same contribution to the output of graph-level representation. In this paper, we develop a novel architecture for extracting an effective graph representation by introducing structured multi-head self-attention in which the attention mechanism consists of three different forms, i.e., node-focused, layer-focused and graph-focused. In order to make full use of the information of graphs, the node-focused self-attention firstly aggregates neighbor node features with a scaled dot-product manner, and then the layer-focused and graph-focused self-attention serve as readout module to measure the importance of different nodes and layers to the model’s output. Moreover, it is able to improve the performance on graph classification tasks by combining these two self-attention mechanisms with base node-level GNNs. The proposed Structured Self-attention Architecture is evaluated on two kinds of graph benchmarks: bioinformatics datasets and social network datasets. Extensive experiments have demonstrated superior performance improvement to existing methods on predictive accuracy." @default.
- W2985111225 created "2019-11-22" @default.
- W2985111225 creator A5000102478 @default.
- W2985111225 creator A5017834338 @default.
- W2985111225 creator A5019560977 @default.
- W2985111225 creator A5087319429 @default.
- W2985111225 creator A5091227928 @default.
- W2985111225 date "2020-04-01" @default.
- W2985111225 modified "2023-10-15" @default.
- W2985111225 title "Structured self-attention architecture for graph-level representation learning" @default.
- W2985111225 cites W1973848216 @default.
- W2985111225 cites W2051998685 @default.
- W2985111225 cites W2054617313 @default.
- W2985111225 cites W2056562706 @default.
- W2985111225 cites W2092750499 @default.
- W2985111225 cites W2116341502 @default.
- W2985111225 cites W2519754777 @default.
- W2985111225 cites W2599837009 @default.
- W2985111225 cites W2777985171 @default.
- W2985111225 cites W2801877145 @default.
- W2985111225 cites W2895979306 @default.
- W2985111225 cites W2903642350 @default.
- W2985111225 cites W2962949934 @default.
- W2985111225 doi "https://doi.org/10.1016/j.patcog.2019.107084" @default.
- W2985111225 hasPublicationYear "2020" @default.
- W2985111225 type Work @default.
- W2985111225 sameAs 2985111225 @default.
- W2985111225 citedByCount "28" @default.
- W2985111225 countsByYear W29851112252019 @default.
- W2985111225 countsByYear W29851112252020 @default.
- W2985111225 countsByYear W29851112252021 @default.
- W2985111225 countsByYear W29851112252022 @default.
- W2985111225 countsByYear W29851112252023 @default.
- W2985111225 crossrefType "journal-article" @default.
- W2985111225 hasAuthorship W2985111225A5000102478 @default.
- W2985111225 hasAuthorship W2985111225A5017834338 @default.
- W2985111225 hasAuthorship W2985111225A5019560977 @default.
- W2985111225 hasAuthorship W2985111225A5087319429 @default.
- W2985111225 hasAuthorship W2985111225A5091227928 @default.
- W2985111225 hasConcept C124101348 @default.
- W2985111225 hasConcept C132525143 @default.
- W2985111225 hasConcept C154945302 @default.
- W2985111225 hasConcept C2993807640 @default.
- W2985111225 hasConcept C41008148 @default.
- W2985111225 hasConcept C59404180 @default.
- W2985111225 hasConcept C70437156 @default.
- W2985111225 hasConcept C80444323 @default.
- W2985111225 hasConceptScore W2985111225C124101348 @default.
- W2985111225 hasConceptScore W2985111225C132525143 @default.
- W2985111225 hasConceptScore W2985111225C154945302 @default.
- W2985111225 hasConceptScore W2985111225C2993807640 @default.
- W2985111225 hasConceptScore W2985111225C41008148 @default.
- W2985111225 hasConceptScore W2985111225C59404180 @default.
- W2985111225 hasConceptScore W2985111225C70437156 @default.
- W2985111225 hasConceptScore W2985111225C80444323 @default.
- W2985111225 hasFunder F4320321001 @default.
- W2985111225 hasFunder F4320335777 @default.
- W2985111225 hasLocation W29851112251 @default.
- W2985111225 hasOpenAccess W2985111225 @default.
- W2985111225 hasPrimaryLocation W29851112251 @default.
- W2985111225 hasRelatedWork W2817444178 @default.
- W2985111225 hasRelatedWork W2914959431 @default.
- W2985111225 hasRelatedWork W2942587884 @default.
- W2985111225 hasRelatedWork W2944724518 @default.
- W2985111225 hasRelatedWork W3099473271 @default.
- W2985111225 hasRelatedWork W3145941582 @default.
- W2985111225 hasRelatedWork W3183761761 @default.
- W2985111225 hasRelatedWork W4296582695 @default.
- W2985111225 hasRelatedWork W4296821948 @default.
- W2985111225 hasRelatedWork W4307933185 @default.
- W2985111225 hasVolume "100" @default.
- W2985111225 isParatext "false" @default.
- W2985111225 isRetracted "false" @default.
- W2985111225 magId "2985111225" @default.
- W2985111225 workType "article" @default.