Understanding the Message Passing in Graph Neural Networks via Power Iteration.
Xue LiYuanzhi ChengPublished in: CoRR (2020)
Keyphrases
- message passing
- neural network
- belief propagation
- distributed systems
- shared memory
- markov random field
- probabilistic inference
- factor graphs
- singly connected
- junction tree
- inference in graphical models
- sum product algorithm
- approximate inference
- tree reweighted
- max product
- power consumption
- matrix multiplication
- lp relaxation
- graphical models
- pairwise