Loop Series and Bethe Variational Bounds in Attractive Graphical Models.
Erik B. SudderthMartin J. WainwrightAlan S. WillskyPublished in: NIPS (2007)
Keyphrases
- graphical models
- free energy
- belief propagation
- approximate inference
- probabilistic model
- message passing
- upper bound
- probabilistic inference
- probabilistic graphical models
- random variables
- loopy belief propagation
- bayesian networks
- variational methods
- belief networks
- marginal probabilities
- markov networks
- map inference
- generalized belief propagation
- structure learning
- exact inference
- lower bound
- conditional random fields
- fixed point
- conditional independence
- image segmentation
- factor graphs
- statistical inference
- energy minimization
- upper and lower bounds
- chain graphs
- optical flow
- generative model
- undirected graphical models
- parameter estimation
- latent variables
- relational dependency networks