Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951659295> ?p ?o ?g. }
- W2951659295 abstract "Recently, graph neural networks (GNNs) have revolutionized the field of graph representation learning through effectively learned node embeddings, and achieved state-of-the-art results in tasks such as node classification and link prediction. However, current GNN methods are inherently flat and do not learn hierarchical representations of graphs---a limitation that is especially problematic for the task of graph classification, where the goal is to predict the label associated with an entire graph. Here we propose DiffPool, a differentiable graph pooling module that can generate hierarchical representations of graphs and can be combined with various graph neural network architectures in an end-to-end fashion. DiffPool learns a differentiable soft cluster assignment for nodes at each layer of a deep GNN, mapping nodes to a set of clusters, which then form the coarsened input for the next GNN layer. Our experimental results show that combining existing GNN methods with DiffPool yields an average improvement of 5-10% accuracy on graph classification benchmarks, compared to all existing pooling approaches, achieving a new state-of-the-art on four out of five benchmark data sets." @default.
- W2951659295 created "2019-06-27" @default.
- W2951659295 creator A5003491365 @default.
- W2951659295 creator A5004542284 @default.
- W2951659295 creator A5009408707 @default.
- W2951659295 creator A5025147185 @default.
- W2951659295 creator A5078337825 @default.
- W2951659295 creator A5091272738 @default.
- W2951659295 date "2018-06-22" @default.
- W2951659295 modified "2023-09-23" @default.
- W2951659295 title "Hierarchical Graph Representation Learning with Differentiable Pooling" @default.
- W2951659295 cites W2046589863 @default.
- W2951659295 cites W2086789804 @default.
- W2951659295 cites W2092750499 @default.
- W2951659295 cites W2099438806 @default.
- W2951659295 cites W2116341502 @default.
- W2951659295 cites W2124824205 @default.
- W2951659295 cites W2135957668 @default.
- W2951659295 cites W2139939114 @default.
- W2951659295 cites W2142498761 @default.
- W2951659295 cites W2147286743 @default.
- W2951659295 cites W2153635508 @default.
- W2951659295 cites W2159156271 @default.
- W2951659295 cites W2173183968 @default.
- W2951659295 cites W2184466087 @default.
- W2951659295 cites W2406128552 @default.
- W2951659295 cites W2558748708 @default.
- W2951659295 cites W2604314403 @default.
- W2951659295 cites W2624431344 @default.
- W2951659295 cites W2751808960 @default.
- W2951659295 cites W2753921001 @default.
- W2951659295 cites W2768308213 @default.
- W2951659295 cites W2788919350 @default.
- W2951659295 cites W2803471865 @default.
- W2951659295 cites W2949117887 @default.
- W2951659295 cites W2950191616 @default.
- W2951659295 cites W2950898568 @default.
- W2951659295 cites W2952254971 @default.
- W2951659295 cites W2963323889 @default.
- W2951659295 cites W2963460103 @default.
- W2951659295 cites W2963464736 @default.
- W2951659295 cites W2963711743 @default.
- W2951659295 cites W2963858333 @default.
- W2951659295 cites W2964311892 @default.
- W2951659295 cites W2964321699 @default.
- W2951659295 hasPublicationYear "2018" @default.
- W2951659295 type Work @default.
- W2951659295 sameAs 2951659295 @default.
- W2951659295 citedByCount "40" @default.
- W2951659295 countsByYear W29516592952018 @default.
- W2951659295 countsByYear W29516592952019 @default.
- W2951659295 countsByYear W29516592952020 @default.
- W2951659295 countsByYear W29516592952021 @default.
- W2951659295 countsByYear W29516592952022 @default.
- W2951659295 countsByYear W29516592952023 @default.
- W2951659295 crossrefType "posted-content" @default.
- W2951659295 hasAuthorship W2951659295A5003491365 @default.
- W2951659295 hasAuthorship W2951659295A5004542284 @default.
- W2951659295 hasAuthorship W2951659295A5009408707 @default.
- W2951659295 hasAuthorship W2951659295A5025147185 @default.
- W2951659295 hasAuthorship W2951659295A5078337825 @default.
- W2951659295 hasAuthorship W2951659295A5091272738 @default.
- W2951659295 hasConcept C119857082 @default.
- W2951659295 hasConcept C132525143 @default.
- W2951659295 hasConcept C13280743 @default.
- W2951659295 hasConcept C134306372 @default.
- W2951659295 hasConcept C154945302 @default.
- W2951659295 hasConcept C185798385 @default.
- W2951659295 hasConcept C202615002 @default.
- W2951659295 hasConcept C205649164 @default.
- W2951659295 hasConcept C33923547 @default.
- W2951659295 hasConcept C41008148 @default.
- W2951659295 hasConcept C70437156 @default.
- W2951659295 hasConcept C80444323 @default.
- W2951659295 hasConceptScore W2951659295C119857082 @default.
- W2951659295 hasConceptScore W2951659295C132525143 @default.
- W2951659295 hasConceptScore W2951659295C13280743 @default.
- W2951659295 hasConceptScore W2951659295C134306372 @default.
- W2951659295 hasConceptScore W2951659295C154945302 @default.
- W2951659295 hasConceptScore W2951659295C185798385 @default.
- W2951659295 hasConceptScore W2951659295C202615002 @default.
- W2951659295 hasConceptScore W2951659295C205649164 @default.
- W2951659295 hasConceptScore W2951659295C33923547 @default.
- W2951659295 hasConceptScore W2951659295C41008148 @default.
- W2951659295 hasConceptScore W2951659295C70437156 @default.
- W2951659295 hasConceptScore W2951659295C80444323 @default.
- W2951659295 hasLocation W29516592951 @default.
- W2951659295 hasOpenAccess W2951659295 @default.
- W2951659295 hasPrimaryLocation W29516592951 @default.
- W2951659295 hasRelatedWork W1662382123 @default.
- W2951659295 hasRelatedWork W2008857988 @default.
- W2951659295 hasRelatedWork W2116341502 @default.
- W2951659295 hasRelatedWork W2142498761 @default.
- W2951659295 hasRelatedWork W2519887557 @default.
- W2951659295 hasRelatedWork W2558460151 @default.
- W2951659295 hasRelatedWork W2606780347 @default.
- W2951659295 hasRelatedWork W2624431344 @default.
- W2951659295 hasRelatedWork W2766453196 @default.
- W2951659295 hasRelatedWork W2788919350 @default.
- W2951659295 hasRelatedWork W2805516822 @default.