Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287251684> ?p ?o ?g. }
Showing items 1 to 51 of
51
with 100 items per page.
- W4287251684 abstract "Graph neural networks have become a staple in problems addressing learning and analysis of data defined over graphs. However, several results suggest an inherent difficulty in extracting better performance by increasing the number of layers. Recent works attribute this to a phenomenon peculiar to the extraction of node features in graph-based tasks, i.e., the need to consider multiple neighborhood sizes at the same time and adaptively tune them. In this paper, we investigate the recently proposed randomly wired architectures in the context of graph neural networks. Instead of building deeper networks by stacking many layers, we prove that employing a randomly-wired architecture can be a more effective way to increase the capacity of the network and obtain richer representations. We show that such architectures behave like an ensemble of paths, which are able to merge contributions from receptive fields of varied size. Moreover, these receptive fields can also be modulated to be wider or narrower through the trainable weights over the paths. We also provide extensive experimental evidence of the superior performance of randomly wired architectures over multiple tasks and four graph convolution definitions, using recent benchmarking frameworks that addresses the reliability of previous testing methodologies." @default.
- W4287251684 created "2022-07-25" @default.
- W4287251684 creator A5034486931 @default.
- W4287251684 creator A5058182410 @default.
- W4287251684 creator A5062328236 @default.
- W4287251684 date "2021-03-29" @default.
- W4287251684 modified "2023-09-23" @default.
- W4287251684 title "RAN-GNNs: breaking the capacity limits of graph neural networks" @default.
- W4287251684 doi "https://doi.org/10.48550/arxiv.2103.15565" @default.
- W4287251684 hasPublicationYear "2021" @default.
- W4287251684 type Work @default.
- W4287251684 citedByCount "0" @default.
- W4287251684 crossrefType "posted-content" @default.
- W4287251684 hasAuthorship W4287251684A5034486931 @default.
- W4287251684 hasAuthorship W4287251684A5058182410 @default.
- W4287251684 hasAuthorship W4287251684A5062328236 @default.
- W4287251684 hasBestOaLocation W42872516841 @default.
- W4287251684 hasConcept C132525143 @default.
- W4287251684 hasConcept C144133560 @default.
- W4287251684 hasConcept C154945302 @default.
- W4287251684 hasConcept C162853370 @default.
- W4287251684 hasConcept C173608175 @default.
- W4287251684 hasConcept C197129107 @default.
- W4287251684 hasConcept C41008148 @default.
- W4287251684 hasConcept C80444323 @default.
- W4287251684 hasConcept C86251818 @default.
- W4287251684 hasConceptScore W4287251684C132525143 @default.
- W4287251684 hasConceptScore W4287251684C144133560 @default.
- W4287251684 hasConceptScore W4287251684C154945302 @default.
- W4287251684 hasConceptScore W4287251684C162853370 @default.
- W4287251684 hasConceptScore W4287251684C173608175 @default.
- W4287251684 hasConceptScore W4287251684C197129107 @default.
- W4287251684 hasConceptScore W4287251684C41008148 @default.
- W4287251684 hasConceptScore W4287251684C80444323 @default.
- W4287251684 hasConceptScore W4287251684C86251818 @default.
- W4287251684 hasLocation W42872516841 @default.
- W4287251684 hasOpenAccess W4287251684 @default.
- W4287251684 hasPrimaryLocation W42872516841 @default.
- W4287251684 hasRelatedWork W1537716277 @default.
- W4287251684 hasRelatedWork W1848359393 @default.
- W4287251684 hasRelatedWork W2127025027 @default.
- W4287251684 hasRelatedWork W2155350564 @default.
- W4287251684 hasRelatedWork W2158198137 @default.
- W4287251684 hasRelatedWork W2369673098 @default.
- W4287251684 hasRelatedWork W2386315983 @default.
- W4287251684 hasRelatedWork W2999063186 @default.
- W4287251684 hasRelatedWork W3009603553 @default.
- W4287251684 hasRelatedWork W4206693874 @default.
- W4287251684 isParatext "false" @default.
- W4287251684 isRetracted "false" @default.
- W4287251684 workType "article" @default.