Matches in SemOpenAlex for { <https://semopenalex.org/work/W3098366174> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3098366174 abstract "Graph neural networks (GNNs) have achieved strong performance in various applications. In the real world, network data is usually formed in a streaming fashion. The distributions of patterns that refer to neighborhood information of nodes may shift over time. The GNN model needs to learn the new patterns that cannot yet be captured. But learning incrementally leads to the catastrophic forgetting problem that historical knowledge is overwritten by newly learned knowledge. Therefore, it is important to train GNN model to learn new patterns and maintain existing patterns simultaneously, which few works focus on. In this paper, we propose a streaming GNN model based on continual learning so that the model is trained incrementally and up-to-date node representations can be obtained at each time step. Firstly, we design an approximation algorithm to detect new coming patterns efficiently based on information propagation. Secondly, we combine two perspectives of data replaying and model regularization for existing pattern consolidation. Specially, a hierarchy-importance sampling strategy for nodes is designed and a weighted regularization term for GNN parameters is derived, achieving greater stability and generalization of knowledge consolidation. Our model is evaluated on real and synthetic data sets and compared with multiple baselines. The results of node classification prove that our model can efficiently update model parameters and achieve comparable performance to model retraining. In addition, we also conduct a case study on the synthetic data, and carry out some specific analysis for each part of our model, illustrating its ability to learn new knowledge and maintain existing knowledge from different perspectives." @default.
- W3098366174 created "2020-11-23" @default.
- W3098366174 creator A5040539235 @default.
- W3098366174 creator A5052258637 @default.
- W3098366174 creator A5086664647 @default.
- W3098366174 creator A5088976879 @default.
- W3098366174 date "2020-10-19" @default.
- W3098366174 modified "2023-10-14" @default.
- W3098366174 title "Streaming Graph Neural Networks via Continual Learning" @default.
- W3098366174 cites W2022322548 @default.
- W3098366174 cites W2119885577 @default.
- W3098366174 cites W2162630660 @default.
- W3098366174 cites W2244109919 @default.
- W3098366174 cites W2560647685 @default.
- W3098366174 cites W2623187518 @default.
- W3098366174 cites W2787927827 @default.
- W3098366174 cites W2803831897 @default.
- W3098366174 cites W2808087697 @default.
- W3098366174 cites W2808908091 @default.
- W3098366174 cites W2950369002 @default.
- W3098366174 cites W2963072899 @default.
- W3098366174 cites W2963076818 @default.
- W3098366174 cites W2963588172 @default.
- W3098366174 cites W2964189064 @default.
- W3098366174 cites W2982701845 @default.
- W3098366174 cites W3104097132 @default.
- W3098366174 cites W3105705953 @default.
- W3098366174 cites W4252973173 @default.
- W3098366174 doi "https://doi.org/10.1145/3340531.3411963" @default.
- W3098366174 hasPublicationYear "2020" @default.
- W3098366174 type Work @default.
- W3098366174 sameAs 3098366174 @default.
- W3098366174 citedByCount "35" @default.
- W3098366174 countsByYear W30983661742020 @default.
- W3098366174 countsByYear W30983661742021 @default.
- W3098366174 countsByYear W30983661742022 @default.
- W3098366174 countsByYear W30983661742023 @default.
- W3098366174 crossrefType "proceedings-article" @default.
- W3098366174 hasAuthorship W3098366174A5040539235 @default.
- W3098366174 hasAuthorship W3098366174A5052258637 @default.
- W3098366174 hasAuthorship W3098366174A5086664647 @default.
- W3098366174 hasAuthorship W3098366174A5088976879 @default.
- W3098366174 hasBestOaLocation W30983661742 @default.
- W3098366174 hasConcept C119857082 @default.
- W3098366174 hasConcept C124101348 @default.
- W3098366174 hasConcept C138885662 @default.
- W3098366174 hasConcept C144133560 @default.
- W3098366174 hasConcept C154945302 @default.
- W3098366174 hasConcept C155202549 @default.
- W3098366174 hasConcept C2776135515 @default.
- W3098366174 hasConcept C2778712577 @default.
- W3098366174 hasConcept C41008148 @default.
- W3098366174 hasConcept C41895202 @default.
- W3098366174 hasConcept C50644808 @default.
- W3098366174 hasConcept C7149132 @default.
- W3098366174 hasConceptScore W3098366174C119857082 @default.
- W3098366174 hasConceptScore W3098366174C124101348 @default.
- W3098366174 hasConceptScore W3098366174C138885662 @default.
- W3098366174 hasConceptScore W3098366174C144133560 @default.
- W3098366174 hasConceptScore W3098366174C154945302 @default.
- W3098366174 hasConceptScore W3098366174C155202549 @default.
- W3098366174 hasConceptScore W3098366174C2776135515 @default.
- W3098366174 hasConceptScore W3098366174C2778712577 @default.
- W3098366174 hasConceptScore W3098366174C41008148 @default.
- W3098366174 hasConceptScore W3098366174C41895202 @default.
- W3098366174 hasConceptScore W3098366174C50644808 @default.
- W3098366174 hasConceptScore W3098366174C7149132 @default.
- W3098366174 hasFunder F4320327720 @default.
- W3098366174 hasLocation W30983661741 @default.
- W3098366174 hasLocation W30983661742 @default.
- W3098366174 hasLocation W30983661743 @default.
- W3098366174 hasOpenAccess W3098366174 @default.
- W3098366174 hasPrimaryLocation W30983661741 @default.
- W3098366174 hasRelatedWork W2118795262 @default.
- W3098366174 hasRelatedWork W2128707566 @default.
- W3098366174 hasRelatedWork W2391375650 @default.
- W3098366174 hasRelatedWork W2961085424 @default.
- W3098366174 hasRelatedWork W3022155277 @default.
- W3098366174 hasRelatedWork W4205701320 @default.
- W3098366174 hasRelatedWork W4306674287 @default.
- W3098366174 hasRelatedWork W4382239683 @default.
- W3098366174 hasRelatedWork W4224009465 @default.
- W3098366174 hasRelatedWork W4287808770 @default.
- W3098366174 isParatext "false" @default.
- W3098366174 isRetracted "false" @default.
- W3098366174 magId "3098366174" @default.
- W3098366174 workType "article" @default.