FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210917077> ?p ?o ?g. }

• W3210917077 abstract "Recent works reveal that feature or label smoothing lies at the core of Graph Neural Networks (GNNs). Concretely, they show feature smoothing combined with simple linear regression achieves comparable performance with the carefully designed GNNs, and a simple MLP model with label smoothing of its prediction can outperform the vanilla GCN. Though an interesting finding, smoothing has not been well understood, especially regarding how to control the extent of smoothness. Intuitively, too small or too large smoothing iterations may cause under-smoothing or over-smoothing and can lead to sub-optimal performance. Moreover, the extent of smoothness is node-specific, depending on its degree and local structure. To this end, we propose a novel algorithm called node-dependent local smoothing (NDLS), which aims to control the smoothness of every node by setting a node-specific smoothing iteration. Specifically, NDLS computes influence scores based on the adjacency matrix and selects the iteration number by setting a threshold on the scores. Once selected, the iteration number can be applied to both feature smoothing and label smoothing. Experimental results demonstrate that NDLS enjoys high accuracy -- state-of-the-art performance on node classifications tasks, flexibility -- can be incorporated with any models, scalability and efficiency -- can support large scale graphs with fast training." @default.
• W3210917077 created "2021-11-08" @default.
• W3210917077 creator A5024993586 @default.
• W3210917077 creator A5042595226 @default.
• W3210917077 creator A5060010918 @default.
• W3210917077 creator A5062357883 @default.
• W3210917077 creator A5064255264 @default.
• W3210917077 creator A5064842058 @default.
• W3210917077 creator A5065994224 @default.
• W3210917077 creator A5077922409 @default.
• W3210917077 date "2021-10-27" @default.
• W3210917077 modified "2023-09-27" @default.
• W3210917077 title "Node Dependent Local Smoothing for Scalable Graph Learning" @default.
• W3210917077 cites W1578099820 @default.
• W3210917077 cites W2132022337 @default.
• W3210917077 cites W2890703109 @default.
• W3210917077 cites W2900763475 @default.
• W3210917077 cites W2916106175 @default.
• W3210917077 cites W2945827377 @default.
• W3210917077 cites W2962767366 @default.
• W3210917077 cites W2963581908 @default.
• W3210917077 cites W2963695795 @default.
• W3210917077 cites W2963858333 @default.
• W3210917077 cites W2964015378 @default.
• W3210917077 cites W2964051675 @default.
• W3210917077 cites W2964114465 @default.
• W3210917077 cites W2966502719 @default.
• W3210917077 cites W2994968268 @default.
• W3210917077 cites W2998008360 @default.
• W3210917077 cites W3002924435 @default.
• W3210917077 cites W3007364240 @default.
• W3210917077 cites W3019011053 @default.
• W3210917077 cites W3030204405 @default.
• W3210917077 cites W3039500550 @default.
• W3210917077 cites W3042770487 @default.
• W3210917077 cites W3045200674 @default.
• W3210917077 cites W3046123837 @default.
• W3210917077 cites W3080510905 @default.
• W3210917077 cites W3097300053 @default.
• W3210917077 cites W3101413764 @default.
• W3210917077 cites W3102476541 @default.
• W3210917077 cites W3104425534 @default.
• W3210917077 cites W3104697628 @default.
• W3210917077 cites W3124006607 @default.
• W3210917077 cites W3147552019 @default.
• W3210917077 cites W3167936733 @default.
• W3210917077 cites W3171733623 @default.
• W3210917077 cites W3189231572 @default.
• W3210917077 cites W3189324775 @default.
• W3210917077 cites W3194434277 @default.
• W3210917077 cites W3207104139 @default.
• W3210917077 hasPublicationYear "2021" @default.
• W3210917077 type Work @default.
• W3210917077 sameAs 3210917077 @default.
• W3210917077 citedByCount "0" @default.
• W3210917077 crossrefType "posted-content" @default.
• W3210917077 hasAuthorship W3210917077A5024993586 @default.
• W3210917077 hasAuthorship W3210917077A5042595226 @default.
• W3210917077 hasAuthorship W3210917077A5060010918 @default.
• W3210917077 hasAuthorship W3210917077A5062357883 @default.
• W3210917077 hasAuthorship W3210917077A5064255264 @default.
• W3210917077 hasAuthorship W3210917077A5064842058 @default.
• W3210917077 hasAuthorship W3210917077A5065994224 @default.
• W3210917077 hasAuthorship W3210917077A5077922409 @default.
• W3210917077 hasBestOaLocation W32109170771 @default.
• W3210917077 hasConcept C102634674 @default.
• W3210917077 hasConcept C11413529 @default.
• W3210917077 hasConcept C127413603 @default.
• W3210917077 hasConcept C132525143 @default.
• W3210917077 hasConcept C134306372 @default.
• W3210917077 hasConcept C138885662 @default.
• W3210917077 hasConcept C154945302 @default.
• W3210917077 hasConcept C180356752 @default.
• W3210917077 hasConcept C2776401178 @default.
• W3210917077 hasConcept C31972630 @default.
• W3210917077 hasConcept C33923547 @default.
• W3210917077 hasConcept C3770464 @default.
• W3210917077 hasConcept C41008148 @default.
• W3210917077 hasConcept C41895202 @default.
• W3210917077 hasConcept C48044578 @default.
• W3210917077 hasConcept C62611344 @default.
• W3210917077 hasConcept C66938386 @default.
• W3210917077 hasConcept C77088390 @default.
• W3210917077 hasConcept C80444323 @default.
• W3210917077 hasConceptScore W3210917077C102634674 @default.
• W3210917077 hasConceptScore W3210917077C11413529 @default.
• W3210917077 hasConceptScore W3210917077C127413603 @default.
• W3210917077 hasConceptScore W3210917077C132525143 @default.
• W3210917077 hasConceptScore W3210917077C134306372 @default.
• W3210917077 hasConceptScore W3210917077C138885662 @default.
• W3210917077 hasConceptScore W3210917077C154945302 @default.
• W3210917077 hasConceptScore W3210917077C180356752 @default.
• W3210917077 hasConceptScore W3210917077C2776401178 @default.
• W3210917077 hasConceptScore W3210917077C31972630 @default.
• W3210917077 hasConceptScore W3210917077C33923547 @default.
• W3210917077 hasConceptScore W3210917077C3770464 @default.
• W3210917077 hasConceptScore W3210917077C41008148 @default.
• W3210917077 hasConceptScore W3210917077C41895202 @default.
• W3210917077 hasConceptScore W3210917077C48044578 @default.
• W3210917077 hasConceptScore W3210917077C62611344 @default.

• next