Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100863344> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W3100863344 endingPage "20448" @default.
- W3100863344 startingPage "20437" @default.
- W3100863344 abstract "Representation learning of graph-structured data is challenging because both graph structure and node features carry important information. Graph Neural Networks (GNNs) provide an expressive way to fuse information from network structure and node features. However, GNNs are prone to adversarial attacks. Here we introduce Graph Information Bottleneck (GIB), an information-theoretic principle that optimally balances expressiveness and robustness of the learned representation of graph-structured data. Inheriting from the general Information Bottleneck (IB), GIB aims to learn the minimal sufficient representation for a given task by maximizing the mutual information between the representation and the target, and simultaneously constraining the mutual information between the representation and the input data. Different from the general IB, GIB regularizes the structural as well as the feature information. We design two sampling algorithms for structural regularization and instantiate the GIB principle with two new models: GIB-Cat and GIB-Bern, and demonstrate the benefits by evaluating the resilience to adversarial attacks. We show that our proposed models are more robust than state-of-the-art graph defense models. GIB-based models empirically achieve up to 31% improvement with adversarial perturbation of the graph structure as well as node features." @default.
- W3100863344 created "2020-11-23" @default.
- W3100863344 creator A5049507829 @default.
- W3100863344 creator A5078152154 @default.
- W3100863344 creator A5086290413 @default.
- W3100863344 creator A5091272738 @default.
- W3100863344 date "2020-10-24" @default.
- W3100863344 modified "2023-09-27" @default.
- W3100863344 title "Graph Information Bottleneck" @default.
- W3100863344 hasPublicationYear "2020" @default.
- W3100863344 type Work @default.
- W3100863344 sameAs 3100863344 @default.
- W3100863344 citedByCount "12" @default.
- W3100863344 countsByYear W31008633442021 @default.
- W3100863344 crossrefType "proceedings-article" @default.
- W3100863344 hasAuthorship W3100863344A5049507829 @default.
- W3100863344 hasAuthorship W3100863344A5078152154 @default.
- W3100863344 hasAuthorship W3100863344A5086290413 @default.
- W3100863344 hasAuthorship W3100863344A5091272738 @default.
- W3100863344 hasConcept C132525143 @default.
- W3100863344 hasConcept C149635348 @default.
- W3100863344 hasConcept C152139883 @default.
- W3100863344 hasConcept C154945302 @default.
- W3100863344 hasConcept C2780513914 @default.
- W3100863344 hasConcept C37736160 @default.
- W3100863344 hasConcept C41008148 @default.
- W3100863344 hasConcept C60008888 @default.
- W3100863344 hasConcept C80444323 @default.
- W3100863344 hasConceptScore W3100863344C132525143 @default.
- W3100863344 hasConceptScore W3100863344C149635348 @default.
- W3100863344 hasConceptScore W3100863344C152139883 @default.
- W3100863344 hasConceptScore W3100863344C154945302 @default.
- W3100863344 hasConceptScore W3100863344C2780513914 @default.
- W3100863344 hasConceptScore W3100863344C37736160 @default.
- W3100863344 hasConceptScore W3100863344C41008148 @default.
- W3100863344 hasConceptScore W3100863344C60008888 @default.
- W3100863344 hasConceptScore W3100863344C80444323 @default.
- W3100863344 hasLocation W31008633441 @default.
- W3100863344 hasOpenAccess W3100863344 @default.
- W3100863344 hasPrimaryLocation W31008633441 @default.
- W3100863344 hasRelatedWork W1647260487 @default.
- W3100863344 hasRelatedWork W2912077575 @default.
- W3100863344 hasRelatedWork W2939638899 @default.
- W3100863344 hasRelatedWork W2955729488 @default.
- W3100863344 hasRelatedWork W2962711740 @default.
- W3100863344 hasRelatedWork W2964015378 @default.
- W3100863344 hasRelatedWork W2972628494 @default.
- W3100863344 hasRelatedWork W2989985603 @default.
- W3100863344 hasRelatedWork W3021801498 @default.
- W3100863344 hasRelatedWork W3033495140 @default.
- W3100863344 hasRelatedWork W3094193403 @default.
- W3100863344 hasRelatedWork W3125118762 @default.
- W3100863344 hasRelatedWork W3128129382 @default.
- W3100863344 hasRelatedWork W3161879162 @default.
- W3100863344 hasRelatedWork W3167650166 @default.
- W3100863344 hasRelatedWork W3173856251 @default.
- W3100863344 hasRelatedWork W3176642478 @default.
- W3100863344 hasRelatedWork W3197064582 @default.
- W3100863344 hasRelatedWork W3199999577 @default.
- W3100863344 hasRelatedWork W3200174181 @default.
- W3100863344 hasVolume "33" @default.
- W3100863344 isParatext "false" @default.
- W3100863344 isRetracted "false" @default.
- W3100863344 magId "3100863344" @default.
- W3100863344 workType "article" @default.