FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3085479328> ?p ?o ?g. }

• W3085479328 abstract "Graph neural networks (GNNs) have achieved superior performance in various applications, but training dedicated GNNs can be costly for large-scale graphs. Some recent work started to study the pre-training of GNNs. However, none of them provide theoretical insights into the design of their frameworks, or clear requirements and guarantees towards their transferability. In this work, we establish a theoretically grounded and practically useful framework for the transfer learning of GNNs. Firstly, we propose a novel view towards the essential graph information and advocate the capturing of it as the goal of transferable GNN training, which motivates the design of EGI (Ego-Graph Information maximization) to analytically achieve this goal. Secondly, when node features are structure-relevant, we conduct an analysis of EGI transferability regarding the difference between the local graph Laplacians of the source and target graphs. We conduct controlled synthetic experiments to directly justify our theoretical conclusions. Comprehensive experiments on two real-world network datasets show consistent results in the analyzed setting of direct-transfering, while those on large-scale knowledge graphs show promising results in the more practical setting of transfering with fine-tuning." @default.
• W3085479328 created "2020-09-21" @default.
• W3085479328 creator A5006897094 @default.
• W3085479328 creator A5010603690 @default.
• W3085479328 creator A5019539533 @default.
• W3085479328 creator A5055199412 @default.
• W3085479328 creator A5069352413 @default.
• W3085479328 creator A5089966579 @default.
• W3085479328 date "2020-09-10" @default.
• W3085479328 modified "2023-09-27" @default.
• W3085479328 title "Transfer Learning of Graph Neural Networks with Ego-graph Information Maximization" @default.
• W3085479328 cites W1533230146 @default.
• W3085479328 cites W1578099820 @default.
• W3085479328 cites W1669094309 @default.
• W3085479328 cites W1854214752 @default.
• W3085479328 cites W2001141328 @default.
• W3085479328 cites W2008620264 @default.
• W3085479328 cites W2022166150 @default.
• W3085479328 cites W2053186076 @default.
• W3085479328 cites W2057685268 @default.
• W3085479328 cites W2102988735 @default.
• W3085479328 cites W2111708605 @default.
• W3085479328 cites W2124637492 @default.
• W3085479328 cites W2126746199 @default.
• W3085479328 cites W2130354913 @default.
• W3085479328 cites W2131953535 @default.
• W3085479328 cites W2153579005 @default.
• W3085479328 cites W2156718197 @default.
• W3085479328 cites W2158787690 @default.
• W3085479328 cites W2194775991 @default.
• W3085479328 cites W2554952599 @default.
• W3085479328 cites W2607500032 @default.
• W3085479328 cites W2753160622 @default.
• W3085479328 cites W2806115886 @default.
• W3085479328 cites W2809435521 @default.
• W3085479328 cites W2811124557 @default.
• W3085479328 cites W2887997457 @default.
• W3085479328 cites W2898475587 @default.
• W3085479328 cites W2906831717 @default.
• W3085479328 cites W2925177113 @default.
• W3085479328 cites W2942315111 @default.
• W3085479328 cites W2947894439 @default.
• W3085479328 cites W2962711740 @default.
• W3085479328 cites W2962756421 @default.
• W3085479328 cites W2962767366 @default.
• W3085479328 cites W2963272802 @default.
• W3085479328 cites W2963341924 @default.
• W3085479328 cites W2963341956 @default.
• W3085479328 cites W2963695795 @default.
• W3085479328 cites W2963782635 @default.
• W3085479328 cites W2963858333 @default.
• W3085479328 cites W2964015378 @default.
• W3085479328 cites W2964121744 @default.
• W3085479328 cites W2964311892 @default.
• W3085479328 cites W2964321699 @default.
• W3085479328 cites W2966121357 @default.
• W3085479328 cites W2971074627 @default.
• W3085479328 cites W2971078096 @default.
• W3085479328 cites W2971081194 @default.
• W3085479328 cites W2994710732 @default.
• W3085479328 cites W2995914187 @default.
• W3085479328 cites W2996604169 @default.
• W3085479328 cites W2997342017 @default.
• W3085479328 cites W3007134945 @default.
• W3085479328 cites W3012644407 @default.
• W3085479328 cites W3012816161 @default.
• W3085479328 cites W3080997787 @default.
• W3085479328 cites W3081963674 @default.
• W3085479328 cites W3099152386 @default.
• W3085479328 cites W3099226596 @default.
• W3085479328 cites W3103284874 @default.
• W3085479328 cites W3103629107 @default.
• W3085479328 cites W3104097132 @default.
• W3085479328 cites W3109650979 @default.
• W3085479328 cites W3114303065 @default.
• W3085479328 cites W3127524795 @default.
• W3085479328 cites W3202514640 @default.
• W3085479328 hasPublicationYear "2020" @default.
• W3085479328 type Work @default.
• W3085479328 sameAs 3085479328 @default.
• W3085479328 citedByCount "3" @default.
• W3085479328 countsByYear W30854793282021 @default.
• W3085479328 countsByYear W30854793282022 @default.
• W3085479328 crossrefType "posted-content" @default.
• W3085479328 hasAuthorship W3085479328A5006897094 @default.
• W3085479328 hasAuthorship W3085479328A5010603690 @default.
• W3085479328 hasAuthorship W3085479328A5019539533 @default.
• W3085479328 hasAuthorship W3085479328A5055199412 @default.
• W3085479328 hasAuthorship W3085479328A5069352413 @default.
• W3085479328 hasAuthorship W3085479328A5089966579 @default.
• W3085479328 hasBestOaLocation W30854793281 @default.
• W3085479328 hasConcept C119857082 @default.
• W3085479328 hasConcept C126255220 @default.
• W3085479328 hasConcept C132525143 @default.
• W3085479328 hasConcept C140331021 @default.
• W3085479328 hasConcept C154945302 @default.
• W3085479328 hasConcept C2776330181 @default.
• W3085479328 hasConcept C33923547 @default.
• W3085479328 hasConcept C41008148 @default.
• W3085479328 hasConcept C50644808 @default.

• next