GNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings.
Matthias FeyJan Eric LenssenFrank WeichertJure LeskovecPublished in: ICML (2021)
Keyphrases
- neural network
- pattern recognition
- directed graph
- graph theory
- fuzzy logic
- random walk
- feed forward
- neural nets
- graph representation
- historical data
- highly scalable
- structured data
- weighted graph
- artificial neural networks
- graph structure
- graph matching
- multi layer
- neural network model
- connected components
- vector space
- hilbert space
- multilayer perceptron
- graph partitioning
- web scale
- dependency graph
- spanning tree
- graph databases
- manifold learning
- complex networks
- back propagation
- bayesian networks