Hybrid Inference with Invertible Neural Networks in Factor Graphs.
Bart van ErpBert de VriesPublished in: EUSIPCO (2022)
Keyphrases
- factor graphs
- neural network
- inference problems
- message passing
- graphical models
- belief propagation
- exact inference
- probabilistic inference
- approximate inference
- latent variables
- probabilistic graphical models
- parameter learning
- back propagation
- artificial neural networks
- markov networks
- distributed systems
- bayesian networks
- random variables
- probabilistic model
- image processing
- markov random field
- model construction
- hidden variables
- graph cuts
- least squares