Matches in SemOpenAlex for { <https://semopenalex.org/work/W3102249364> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W3102249364 endingPage "18930" @default.
- W3102249364 startingPage "18917" @default.
- W3102249364 abstract "It is known that the current graph neural networks (GNNs) are difficult to make themselves deep due to the problem known as over-smoothing. Multi-scale GNNs are a promising approach for mitigating the over-smoothing problem. However, there is little explanation of why it works empirically from the viewpoint of learning theory. In this study, we derive the optimization and generalization guarantees of transductive learning algorithms that include multi-scale GNNs. Using the boosting theory, we prove the convergence of the training error under weak learning-type conditions. By combining it with generalization gap bounds in terms of transductive Rademacher complexity, we show that a test error bound of a specific type of multi-scale GNNs that decreases corresponding to the number of node aggregations under some conditions. Our results offer theoretical explanations for the effectiveness of the multi-scale structure against the over-smoothing problem. We apply boosting algorithms to the training of multi-scale GNNs for real-world node prediction tasks. We confirm that its performance is comparable to existing GNNs, and the practical behaviors are consistent with theoretical observations. Code is available at this https URL." @default.
- W3102249364 created "2020-11-23" @default.
- W3102249364 creator A5010759340 @default.
- W3102249364 creator A5078812767 @default.
- W3102249364 date "2020-01-01" @default.
- W3102249364 modified "2023-10-18" @default.
- W3102249364 title "Optimization and Generalization Analysis of Transduction through Gradient Boosting and Application to Multi-scale Graph Neural Networks" @default.
- W3102249364 hasPublicationYear "2020" @default.
- W3102249364 type Work @default.
- W3102249364 sameAs 3102249364 @default.
- W3102249364 citedByCount "6" @default.
- W3102249364 countsByYear W31022493642019 @default.
- W3102249364 countsByYear W31022493642021 @default.
- W3102249364 crossrefType "proceedings-article" @default.
- W3102249364 hasAuthorship W3102249364A5010759340 @default.
- W3102249364 hasAuthorship W3102249364A5078812767 @default.
- W3102249364 hasConcept C119857082 @default.
- W3102249364 hasConcept C132525143 @default.
- W3102249364 hasConcept C134306372 @default.
- W3102249364 hasConcept C154945302 @default.
- W3102249364 hasConcept C169258074 @default.
- W3102249364 hasConcept C177148314 @default.
- W3102249364 hasConcept C31972630 @default.
- W3102249364 hasConcept C33923547 @default.
- W3102249364 hasConcept C3770464 @default.
- W3102249364 hasConcept C41008148 @default.
- W3102249364 hasConcept C46686674 @default.
- W3102249364 hasConcept C50644808 @default.
- W3102249364 hasConcept C5465570 @default.
- W3102249364 hasConcept C70153297 @default.
- W3102249364 hasConcept C80444323 @default.
- W3102249364 hasConceptScore W3102249364C119857082 @default.
- W3102249364 hasConceptScore W3102249364C132525143 @default.
- W3102249364 hasConceptScore W3102249364C134306372 @default.
- W3102249364 hasConceptScore W3102249364C154945302 @default.
- W3102249364 hasConceptScore W3102249364C169258074 @default.
- W3102249364 hasConceptScore W3102249364C177148314 @default.
- W3102249364 hasConceptScore W3102249364C31972630 @default.
- W3102249364 hasConceptScore W3102249364C33923547 @default.
- W3102249364 hasConceptScore W3102249364C3770464 @default.
- W3102249364 hasConceptScore W3102249364C41008148 @default.
- W3102249364 hasConceptScore W3102249364C46686674 @default.
- W3102249364 hasConceptScore W3102249364C50644808 @default.
- W3102249364 hasConceptScore W3102249364C5465570 @default.
- W3102249364 hasConceptScore W3102249364C70153297 @default.
- W3102249364 hasConceptScore W3102249364C80444323 @default.
- W3102249364 hasLocation W31022493641 @default.
- W3102249364 hasOpenAccess W3102249364 @default.
- W3102249364 hasPrimaryLocation W31022493641 @default.
- W3102249364 hasRelatedWork W2751825796 @default.
- W3102249364 hasRelatedWork W2786915849 @default.
- W3102249364 hasRelatedWork W2789088344 @default.
- W3102249364 hasRelatedWork W2804052888 @default.
- W3102249364 hasRelatedWork W2805880533 @default.
- W3102249364 hasRelatedWork W2898221481 @default.
- W3102249364 hasRelatedWork W2963691249 @default.
- W3102249364 hasRelatedWork W2964015378 @default.
- W3102249364 hasRelatedWork W2978103562 @default.
- W3102249364 hasRelatedWork W2987511773 @default.
- W3102249364 hasRelatedWork W2996051514 @default.
- W3102249364 hasRelatedWork W2999913149 @default.
- W3102249364 hasRelatedWork W3082378091 @default.
- W3102249364 hasRelatedWork W3097191881 @default.
- W3102249364 hasRelatedWork W3098708496 @default.
- W3102249364 hasRelatedWork W3100823189 @default.
- W3102249364 hasRelatedWork W3153800554 @default.
- W3102249364 hasRelatedWork W3162265811 @default.
- W3102249364 hasRelatedWork W3210221883 @default.
- W3102249364 hasRelatedWork W3096979466 @default.
- W3102249364 hasVolume "33" @default.
- W3102249364 isParatext "false" @default.
- W3102249364 isRetracted "false" @default.
- W3102249364 magId "3102249364" @default.
- W3102249364 workType "article" @default.