FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964876045> ?p ?o ?g. }

• W2964876045 endingPage "475" @default.
• W2964876045 startingPage "465" @default.
• W2964876045 abstract "In a graph convolutional network, we assume that the graph $G$ is generated wrt some observation noise. During learning, we make small random perturbations $Delta{}G$ of the graph and try to improve generalization. Based on quantum information geometry, $Delta{}G$ can be characterized by the eigendecomposition of the graph Laplacian matrix. We try to minimize the loss wrt the perturbed $G+Delta{G}$ while making $Delta{G}$ to be effective in terms of the Fisher information of the neural network. Our proposed model can consistently improve graph convolutional networks on semi-supervised node classification tasks with reasonable computational overhead. We present three different geometries on the manifold of graphs: the intrinsic geometry measures the information theoretic dynamics of a graph; the extrinsic geometry characterizes how such dynamics can affect externally a graph neural network; the embedding geometry is for measuring node embeddings. These new analytical tools are useful in developing a good understanding of graph neural networks and fostering new techniques." @default.
• W2964876045 created "2019-08-13" @default.
• W2964876045 creator A5002212263 @default.
• W2964876045 creator A5065058975 @default.
• W2964876045 creator A5068048142 @default.
• W2964876045 date "2019-01-01" @default.
• W2964876045 modified "2023-10-08" @default.
• W2964876045 title "Fisher-Bures Adversary Graph Convolutional Networks." @default.
• W2964876045 cites W1501856433 @default.
• W2964876045 cites W1959608418 @default.
• W2964876045 cites W1994618660 @default.
• W2964876045 cites W2024974438 @default.
• W2964876045 cites W2099471712 @default.
• W2964876045 cites W2102714712 @default.
• W2964876045 cites W2107397716 @default.
• W2964876045 cites W2114826682 @default.
• W2964876045 cites W2116341502 @default.
• W2964876045 cites W2130707516 @default.
• W2964876045 cites W2136739711 @default.
• W2964876045 cites W2177697562 @default.
• W2964876045 cites W2187013920 @default.
• W2964876045 cites W2243397390 @default.
• W2964876045 cites W2558460151 @default.
• W2964876045 cites W2624431344 @default.
• W2964876045 cites W2739741052 @default.
• W2964876045 cites W2784814091 @default.
• W2964876045 cites W2804057010 @default.
• W2964876045 cites W2888674590 @default.
• W2964876045 cites W2893944917 @default.
• W2964876045 cites W2897355816 @default.
• W2964876045 cites W2900470550 @default.
• W2964876045 cites W2916106175 @default.
• W2964876045 cites W2949578333 @default.
• W2964876045 cites W2949865801 @default.
• W2964876045 cites W2962739609 @default.
• W2964876045 cites W2962756421 @default.
• W2964876045 cites W2963312446 @default.
• W2964876045 cites W2963581908 @default.
• W2964876045 cites W2963671728 @default.
• W2964876045 cites W2964259376 @default.
• W2964876045 cites W2964309400 @default.
• W2964876045 cites W2964311892 @default.
• W2964876045 cites W3100848837 @default.
• W2964876045 cites W597395834 @default.
• W2964876045 hasPublicationYear "2019" @default.
• W2964876045 type Work @default.
• W2964876045 sameAs 2964876045 @default.
• W2964876045 citedByCount "4" @default.
• W2964876045 countsByYear W29648760452019 @default.
• W2964876045 countsByYear W29648760452020 @default.
• W2964876045 countsByYear W29648760452021 @default.
• W2964876045 crossrefType "proceedings-article" @default.
• W2964876045 hasAuthorship W2964876045A5002212263 @default.
• W2964876045 hasAuthorship W2964876045A5065058975 @default.
• W2964876045 hasAuthorship W2964876045A5068048142 @default.
• W2964876045 hasConcept C109546454 @default.
• W2964876045 hasConcept C11413529 @default.
• W2964876045 hasConcept C12520029 @default.
• W2964876045 hasConcept C132525143 @default.
• W2964876045 hasConcept C154945302 @default.
• W2964876045 hasConcept C180356752 @default.
• W2964876045 hasConcept C195065555 @default.
• W2964876045 hasConcept C203776342 @default.
• W2964876045 hasConcept C22149727 @default.
• W2964876045 hasConcept C2524010 @default.
• W2964876045 hasConcept C30609935 @default.
• W2964876045 hasConcept C33923547 @default.
• W2964876045 hasConcept C41008148 @default.
• W2964876045 hasConcept C80444323 @default.
• W2964876045 hasConcept C81363708 @default.
• W2964876045 hasConceptScore W2964876045C109546454 @default.
• W2964876045 hasConceptScore W2964876045C11413529 @default.
• W2964876045 hasConceptScore W2964876045C12520029 @default.
• W2964876045 hasConceptScore W2964876045C132525143 @default.
• W2964876045 hasConceptScore W2964876045C154945302 @default.
• W2964876045 hasConceptScore W2964876045C180356752 @default.
• W2964876045 hasConceptScore W2964876045C195065555 @default.
• W2964876045 hasConceptScore W2964876045C203776342 @default.
• W2964876045 hasConceptScore W2964876045C22149727 @default.
• W2964876045 hasConceptScore W2964876045C2524010 @default.
• W2964876045 hasConceptScore W2964876045C30609935 @default.
• W2964876045 hasConceptScore W2964876045C33923547 @default.
• W2964876045 hasConceptScore W2964876045C41008148 @default.
• W2964876045 hasConceptScore W2964876045C80444323 @default.
• W2964876045 hasConceptScore W2964876045C81363708 @default.
• W2964876045 hasLocation W29648760451 @default.
• W2964876045 hasOpenAccess W2964876045 @default.
• W2964876045 hasPrimaryLocation W29648760451 @default.
• W2964876045 hasRelatedWork W1581695321 @default.
• W2964876045 hasRelatedWork W2032261599 @default.
• W2964876045 hasRelatedWork W2154136271 @default.
• W2964876045 hasRelatedWork W2487418985 @default.
• W2964876045 hasRelatedWork W2612384513 @default.
• W2964876045 hasRelatedWork W2739222904 @default.
• W2964876045 hasRelatedWork W2748347917 @default.
• W2964876045 hasRelatedWork W2800334336 @default.
• W2964876045 hasRelatedWork W2801603084 @default.
• W2964876045 hasRelatedWork W2885478882 @default.

• next