GMLP: Building Scalable and Flexible Graph Neural Networks with Feature-Message Passing.
Wentao ZhangYu ShenZheyu LinYang LiXiaosen LiWen OuyangYangyu TaoZhi YangBin CuiPublished in: CoRR (2021)
Keyphrases
- message passing
- neural network
- belief propagation
- probabilistic inference
- approximate inference
- shared memory
- distributed systems
- inference in graphical models
- junction tree
- markov random field
- max product
- factor graphs
- sum product algorithm
- singly connected
- distributed shared memory
- graphical models
- multilayer perceptron
- tree reweighted
- ldpc codes
- lp relaxation
- image features
- feature vectors
- preprocessing
- d objects
- sum product
- bayesian networks