FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048798735> ?p ?o ?g. }

• W3048798735 abstract "Recently, graph neural networks (GNNs) become a principal research direction to learn low-dimensional continuous embeddings of nodes and graphs to predict node and graph labels, respectively. However, Euclidean embeddings have high distortion when using GNNs to model complex graphs such as social networks. Furthermore, existing GNNs are not very efficient with the high number of model parameters when increasing the number of hidden layers. Therefore, we move beyond the Euclidean space to a hyper-complex vector space to improve graph representation quality and reduce the number of model parameters. To this end, we propose quaternion graph neural networks (QGNN) to generalize GCNs within the Quaternion space to learn quaternion embeddings for nodes and graphs. The Quaternion space, a hyper-complex vector space, provides highly meaningful computations through Hamilton product compared to the Euclidean and complex vector spaces. As a result, our QGNN can reduce the model size up to four times and enhance learning better graph representations. Experimental results show that the proposed QGNN produces state-of-the-art accuracies on a range of well-known benchmark datasets for three downstream tasks, including graph classification, semi-supervised node classification, and text (node) classification. Our code is available at: this https URL" @default.
• W3048798735 created "2020-08-18" @default.
• W3048798735 creator A5036447132 @default.
• W3048798735 creator A5067423247 @default.
• W3048798735 creator A5086752305 @default.
• W3048798735 date "2020-08-12" @default.
• W3048798735 modified "2023-09-27" @default.
• W3048798735 title "Quaternion Graph Neural Networks" @default.
• W3048798735 cites W1533230146 @default.
• W3048798735 cites W1989286669 @default.
• W3048798735 cites W2008857988 @default.
• W3048798735 cites W2018049970 @default.
• W3048798735 cites W2116341502 @default.
• W3048798735 cites W2127426251 @default.
• W3048798735 cites W2127795553 @default.
• W3048798735 cites W2128407051 @default.
• W3048798735 cites W2142498761 @default.
• W3048798735 cites W2153959628 @default.
• W3048798735 cites W2157331557 @default.
• W3048798735 cites W2159156271 @default.
• W3048798735 cites W2187089797 @default.
• W3048798735 cites W2247119764 @default.
• W3048798735 cites W2250184916 @default.
• W3048798735 cites W2250539671 @default.
• W3048798735 cites W2604314403 @default.
• W3048798735 cites W2624431344 @default.
• W3048798735 cites W2755092149 @default.
• W3048798735 cites W2788919350 @default.
• W3048798735 cites W2803303416 @default.
• W3048798735 cites W2803471865 @default.
• W3048798735 cites W2811124557 @default.
• W3048798735 cites W2888677580 @default.
• W3048798735 cites W2894977338 @default.
• W3048798735 cites W2899379687 @default.
• W3048798735 cites W2905224888 @default.
• W3048798735 cites W2906340703 @default.
• W3048798735 cites W2909882940 @default.
• W3048798735 cites W2918342466 @default.
• W3048798735 cites W2942026896 @default.
• W3048798735 cites W2944538680 @default.
• W3048798735 cites W2947926079 @default.
• W3048798735 cites W2952254971 @default.
• W3048798735 cites W2952383053 @default.
• W3048798735 cites W2962711740 @default.
• W3048798735 cites W2962767366 @default.
• W3048798735 cites W2963230471 @default.
• W3048798735 cites W2963235055 @default.
• W3048798735 cites W2963312446 @default.
• W3048798735 cites W2963403868 @default.
• W3048798735 cites W2963432357 @default.
• W3048798735 cites W2963512530 @default.
• W3048798735 cites W2963719423 @default.
• W3048798735 cites W2964015378 @default.
• W3048798735 cites W2964114465 @default.
• W3048798735 cites W2964116313 @default.
• W3048798735 cites W2964121744 @default.
• W3048798735 cites W2964124573 @default.
• W3048798735 cites W2964145825 @default.
• W3048798735 cites W2964308564 @default.
• W3048798735 cites W2966878259 @default.
• W3048798735 cites W2970474218 @default.
• W3048798735 cites W2970493342 @default.
• W3048798735 cites W2970726356 @default.
• W3048798735 cites W2995448904 @default.
• W3048798735 cites W2995509042 @default.
• W3048798735 cites W3036796504 @default.
• W3048798735 cites W3087818399 @default.
• W3048798735 cites W3093499132 @default.
• W3048798735 cites W3099836348 @default.
• W3048798735 cites W3103296573 @default.
• W3048798735 cites W3164306231 @default.
• W3048798735 cites W67413104 @default.
• W3048798735 cites W2900252255 @default.
• W3048798735 cites W2970757764 @default.
• W3048798735 hasPublicationYear "2020" @default.
• W3048798735 type Work @default.
• W3048798735 sameAs 3048798735 @default.
• W3048798735 citedByCount "3" @default.
• W3048798735 countsByYear W30487987352021 @default.
• W3048798735 crossrefType "posted-content" @default.
• W3048798735 hasAuthorship W3048798735A5036447132 @default.
• W3048798735 hasAuthorship W3048798735A5067423247 @default.
• W3048798735 hasAuthorship W3048798735A5086752305 @default.
• W3048798735 hasConcept C11413529 @default.
• W3048798735 hasConcept C114614502 @default.
• W3048798735 hasConcept C120174047 @default.
• W3048798735 hasConcept C132525143 @default.
• W3048798735 hasConcept C13336665 @default.
• W3048798735 hasConcept C154945302 @default.
• W3048798735 hasConcept C186450821 @default.
• W3048798735 hasConcept C200127275 @default.
• W3048798735 hasConcept C202444582 @default.
• W3048798735 hasConcept C2524010 @default.
• W3048798735 hasConcept C33923547 @default.
• W3048798735 hasConcept C41008148 @default.
• W3048798735 hasConcept C50644808 @default.
• W3048798735 hasConcept C80444323 @default.
• W3048798735 hasConceptScore W3048798735C11413529 @default.
• W3048798735 hasConceptScore W3048798735C114614502 @default.
• W3048798735 hasConceptScore W3048798735C120174047 @default.

• next