Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899379687> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2899379687 abstract "Recent advances in representation learning on graphs, mainly leveraging graph convolutional networks, have brought a substantial improvement on many graph-based benchmark tasks. While novel approaches to learning node embeddings are highly suitable for node classification and link prediction, their application to graph classification (predicting a single label for the entire graph) remains mostly rudimentary, typically using a single global pooling step to aggregate node features or a hand-designed, fixed heuristic for hierarchical coarsening of the graph structure. An important step towards ameliorating this is differentiable graph coarsening---the ability to reduce the size of the graph in an adaptive, data-dependent manner within a graph neural network pipeline, analogous to image downsampling within CNNs. However, the previous prominent approach to pooling has quadratic memory requirements during training and is therefore not scalable to large graphs. Here we combine several recent advances in graph neural network design to demonstrate that competitive hierarchical graph classification results are possible without sacrificing sparsity. Our results are verified on several established graph classification benchmarks, and highlight an important direction for future research in graph-based neural networks." @default.
- W2899379687 created "2018-11-09" @default.
- W2899379687 creator A5008869927 @default.
- W2899379687 creator A5025691113 @default.
- W2899379687 creator A5056748708 @default.
- W2899379687 creator A5072208718 @default.
- W2899379687 creator A5086533240 @default.
- W2899379687 date "2018-11-03" @default.
- W2899379687 modified "2023-10-01" @default.
- W2899379687 title "Towards Sparse Hierarchical Graph Classifiers." @default.
- W2899379687 cites W1522301498 @default.
- W2899379687 cites W1662382123 @default.
- W2899379687 cites W2194775991 @default.
- W2899379687 cites W2244807774 @default.
- W2899379687 cites W2406128552 @default.
- W2899379687 cites W2519887557 @default.
- W2899379687 cites W2558460151 @default.
- W2899379687 cites W2624431344 @default.
- W2899379687 cites W2804057010 @default.
- W2899379687 cites W2805516822 @default.
- W2899379687 cites W2809146571 @default.
- W2899379687 cites W2894175828 @default.
- W2899379687 cites W2899771611 @default.
- W2899379687 cites W2907101105 @default.
- W2899379687 cites W2950191616 @default.
- W2899379687 cites W2952254971 @default.
- W2899379687 cites W2963460103 @default.
- W2899379687 cites W2964321699 @default.
- W2899379687 cites W3143219376 @default.
- W2899379687 hasPublicationYear "2018" @default.
- W2899379687 type Work @default.
- W2899379687 sameAs 2899379687 @default.
- W2899379687 citedByCount "59" @default.
- W2899379687 countsByYear W28993796872018 @default.
- W2899379687 countsByYear W28993796872019 @default.
- W2899379687 countsByYear W28993796872020 @default.
- W2899379687 countsByYear W28993796872021 @default.
- W2899379687 countsByYear W28993796872022 @default.
- W2899379687 crossrefType "posted-content" @default.
- W2899379687 hasAuthorship W2899379687A5008869927 @default.
- W2899379687 hasAuthorship W2899379687A5025691113 @default.
- W2899379687 hasAuthorship W2899379687A5056748708 @default.
- W2899379687 hasAuthorship W2899379687A5072208718 @default.
- W2899379687 hasAuthorship W2899379687A5086533240 @default.
- W2899379687 hasConcept C119857082 @default.
- W2899379687 hasConcept C132525143 @default.
- W2899379687 hasConcept C134306372 @default.
- W2899379687 hasConcept C154945302 @default.
- W2899379687 hasConcept C202615002 @default.
- W2899379687 hasConcept C33923547 @default.
- W2899379687 hasConcept C41008148 @default.
- W2899379687 hasConcept C48044578 @default.
- W2899379687 hasConcept C70437156 @default.
- W2899379687 hasConcept C77088390 @default.
- W2899379687 hasConcept C80444323 @default.
- W2899379687 hasConceptScore W2899379687C119857082 @default.
- W2899379687 hasConceptScore W2899379687C132525143 @default.
- W2899379687 hasConceptScore W2899379687C134306372 @default.
- W2899379687 hasConceptScore W2899379687C154945302 @default.
- W2899379687 hasConceptScore W2899379687C202615002 @default.
- W2899379687 hasConceptScore W2899379687C33923547 @default.
- W2899379687 hasConceptScore W2899379687C41008148 @default.
- W2899379687 hasConceptScore W2899379687C48044578 @default.
- W2899379687 hasConceptScore W2899379687C70437156 @default.
- W2899379687 hasConceptScore W2899379687C77088390 @default.
- W2899379687 hasConceptScore W2899379687C80444323 @default.
- W2899379687 hasLocation W28993796871 @default.
- W2899379687 hasOpenAccess W2899379687 @default.
- W2899379687 hasPrimaryLocation W28993796871 @default.
- W2899379687 hasRelatedWork W2116341502 @default.
- W2899379687 hasRelatedWork W2142498761 @default.
- W2899379687 hasRelatedWork W2558460151 @default.
- W2899379687 hasRelatedWork W2558748708 @default.
- W2899379687 hasRelatedWork W2606202972 @default.
- W2899379687 hasRelatedWork W2606780347 @default.
- W2899379687 hasRelatedWork W2788919350 @default.
- W2899379687 hasRelatedWork W2811124557 @default.
- W2899379687 hasRelatedWork W2918342466 @default.
- W2899379687 hasRelatedWork W2952832237 @default.
- W2899379687 hasRelatedWork W2962711740 @default.
- W2899379687 hasRelatedWork W2962767366 @default.
- W2899379687 hasRelatedWork W2963175980 @default.
- W2899379687 hasRelatedWork W2963858333 @default.
- W2899379687 hasRelatedWork W2964015378 @default.
- W2899379687 hasRelatedWork W2964113829 @default.
- W2899379687 hasRelatedWork W2964121744 @default.
- W2899379687 hasRelatedWork W2964311892 @default.
- W2899379687 hasRelatedWork W2964321699 @default.
- W2899379687 hasRelatedWork W4210257598 @default.
- W2899379687 isParatext "false" @default.
- W2899379687 isRetracted "false" @default.
- W2899379687 magId "2899379687" @default.
- W2899379687 workType "article" @default.