Deep learning via message passing algorithms based on belief propagation.
Carlo LucibelloFabrizio PittorinoGabriele PeruginiRiccardo ZecchinaPublished in: Mach. Learn. Sci. Technol. (2022)
Keyphrases
- message passing
- belief propagation
- graphical models
- max product
- loopy belief propagation
- deep learning
- markov random field
- inference in graphical models
- approximate inference
- clique potentials
- factor graphs
- stereo matching
- graph cuts
- probabilistic inference
- markov networks
- energy function
- distributed systems
- pairwise
- generalized belief propagation
- tree reweighted
- free energy
- energy minimization
- learning algorithm
- decoding algorithm
- sum product
- belief networks
- reinforcement learning
- similarity measure
- machine learning