Uncertainty-Confidence Fused Pseudo-labeling for Graph Neural Networks.
Pingjiang LongZihao JianXiangrong LiuPublished in: PRCV (9) (2023)
Keyphrases
- neural network
- label propagation
- graph theory
- reachability queries
- directed graph
- pattern recognition
- graph representation
- artificial neural networks
- directed acyclic graph
- connected components
- multi layer
- neural nets
- uncertain data
- neural network model
- image segmentation
- data fusion
- confidence level
- fuzzy logic
- active learning
- undirected graph
- graph model
- graph based algorithm
- genetic algorithm
- graph structure
- graph matching
- class labels
- structured data
- random walk
- learning algorithm
- bipartite graph
- recurrent neural networks
- graph mining
- feed forward
- unsupervised learning
- fusion process
- probability measure