Predicting Global Label Relationship Matrix for Graph Neural Networks under Heterophily.
Langzhang LiangXiangjing HuZenglin XuZixing SongIrwin KingPublished in: NeurIPS (2023)
Keyphrases
- neural network
- adjacency matrix
- random walk
- pattern recognition
- graph theory
- graph representation
- label propagation
- structured data
- graph matching
- feed forward
- hopfield neural network
- graph model
- graph structure
- directed graph
- eigenvalues and eigenvectors
- fuzzy logic
- genetic algorithm
- graph clustering
- global consistency
- matrix representation
- graph databases
- low rank and sparse
- graph mining
- neural nets
- connected components
- covariance matrix
- fault diagnosis
- artificial neural networks
- weighted graph
- spanning tree
- singular value decomposition
- multi label
- self organizing maps
- mutual reinforcement
- back propagation