Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287177528> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4287177528 abstract "Most existing graph neural networks (GNNs) learn node embeddings using the framework of message passing and aggregation. Such GNNs are incapable of learning relative positions between graph nodes within a graph. To empower GNNs with the awareness of node positions, some nodes are set as anchors. Then, using the distances from a node to the anchors, GNNs can infer relative positions between nodes. However, P-GNNs arbitrarily select anchors, leading to compromising position-awareness and feature extraction. To eliminate this compromise, we demonstrate that selecting evenly distributed and asymmetric anchors is essential. On the other hand, we show that choosing anchors that can aggregate embeddings of all the nodes within a graph is NP-hard. Therefore, devising efficient optimal algorithms in a deterministic approach is practically not feasible. To ensure position-awareness and bypass NP-completeness, we propose Position-Sensing Graph Neural Networks (PSGNNs), learning how to choose anchors in a back-propagatable fashion. Experiments verify the effectiveness of PSGNNs against state-of-the-art GNNs, substantially improving performance on various synthetic and real-world graph datasets while enjoying stable scalability. Specifically, PSGNNs on average boost AUC more than 14% for pairwise node classification and 18% for link prediction over the existing state-of-the-art position-aware methods. Our source code is publicly available at: https://github.com/ZhenyueQin/PSGNN" @default.
- W4287177528 created "2022-07-25" @default.
- W4287177528 creator A5023363049 @default.
- W4287177528 creator A5024398480 @default.
- W4287177528 creator A5025229672 @default.
- W4287177528 creator A5029093676 @default.
- W4287177528 creator A5030379402 @default.
- W4287177528 creator A5084633048 @default.
- W4287177528 date "2021-05-24" @default.
- W4287177528 modified "2023-09-28" @default.
- W4287177528 title "Position-Sensing Graph Neural Networks: Proactively Learning Nodes Relative Positions" @default.
- W4287177528 hasPublicationYear "2021" @default.
- W4287177528 type Work @default.
- W4287177528 citedByCount "0" @default.
- W4287177528 crossrefType "posted-content" @default.
- W4287177528 hasAuthorship W4287177528A5023363049 @default.
- W4287177528 hasAuthorship W4287177528A5024398480 @default.
- W4287177528 hasAuthorship W4287177528A5025229672 @default.
- W4287177528 hasAuthorship W4287177528A5029093676 @default.
- W4287177528 hasAuthorship W4287177528A5030379402 @default.
- W4287177528 hasAuthorship W4287177528A5084633048 @default.
- W4287177528 hasBestOaLocation W42871775281 @default.
- W4287177528 hasConcept C119857082 @default.
- W4287177528 hasConcept C124101348 @default.
- W4287177528 hasConcept C127413603 @default.
- W4287177528 hasConcept C132525143 @default.
- W4287177528 hasConcept C154945302 @default.
- W4287177528 hasConcept C177264268 @default.
- W4287177528 hasConcept C184898388 @default.
- W4287177528 hasConcept C199360897 @default.
- W4287177528 hasConcept C41008148 @default.
- W4287177528 hasConcept C48044578 @default.
- W4287177528 hasConcept C50644808 @default.
- W4287177528 hasConcept C59404180 @default.
- W4287177528 hasConcept C62611344 @default.
- W4287177528 hasConcept C66938386 @default.
- W4287177528 hasConcept C77088390 @default.
- W4287177528 hasConcept C80444323 @default.
- W4287177528 hasConceptScore W4287177528C119857082 @default.
- W4287177528 hasConceptScore W4287177528C124101348 @default.
- W4287177528 hasConceptScore W4287177528C127413603 @default.
- W4287177528 hasConceptScore W4287177528C132525143 @default.
- W4287177528 hasConceptScore W4287177528C154945302 @default.
- W4287177528 hasConceptScore W4287177528C177264268 @default.
- W4287177528 hasConceptScore W4287177528C184898388 @default.
- W4287177528 hasConceptScore W4287177528C199360897 @default.
- W4287177528 hasConceptScore W4287177528C41008148 @default.
- W4287177528 hasConceptScore W4287177528C48044578 @default.
- W4287177528 hasConceptScore W4287177528C50644808 @default.
- W4287177528 hasConceptScore W4287177528C59404180 @default.
- W4287177528 hasConceptScore W4287177528C62611344 @default.
- W4287177528 hasConceptScore W4287177528C66938386 @default.
- W4287177528 hasConceptScore W4287177528C77088390 @default.
- W4287177528 hasConceptScore W4287177528C80444323 @default.
- W4287177528 hasLocation W42871775281 @default.
- W4287177528 hasOpenAccess W4287177528 @default.
- W4287177528 hasPrimaryLocation W42871775281 @default.
- W4287177528 hasRelatedWork W10763751 @default.
- W4287177528 hasRelatedWork W11244355 @default.
- W4287177528 hasRelatedWork W12712126 @default.
- W4287177528 hasRelatedWork W13536281 @default.
- W4287177528 hasRelatedWork W1549441 @default.
- W4287177528 hasRelatedWork W3990674 @default.
- W4287177528 hasRelatedWork W4771408 @default.
- W4287177528 hasRelatedWork W621929 @default.
- W4287177528 hasRelatedWork W7842670 @default.
- W4287177528 hasRelatedWork W8198582 @default.
- W4287177528 isParatext "false" @default.
- W4287177528 isRetracted "false" @default.
- W4287177528 workType "article" @default.