Semi-Supervised Node Classification on Graphs: Markov Random Fields vs. Graph Neural Networks.
Binghui WangJinyuan JiaNeil Zhenqiang GongPublished in: CoRR (2020)
Keyphrases
- markov random field
- graph structure
- semi supervised
- neural network
- graph construction
- semi supervised classification
- directed graph
- graphical structure
- pattern recognition
- graph cuts
- finding the shortest path
- supervised learning
- undirected graph
- energy function
- pairwise
- belief propagation
- higher order
- parameter estimation
- weighted graph
- mrf model
- min cut
- discriminative random fields
- semi supervised learning
- energy minimization
- random fields
- maximum a posteriori
- image segmentation
- image restoration
- potts model
- potential functions
- machine learning
- image processing
- low level vision
- edge weights
- graph partitioning
- graphical models
- graph model
- random graphs
- markov networks
- unlabeled data
- message passing
- random walk
- loopy belief propagation
- training set
- probabilistic model
- spanning tree
- partition function
- text classification
- max flow
- denoising
- map estimation
- computer vision
- conditional random fields