Minimum Free Energy Principle for Constraint-Based Learning Bayesian Networks.
Takashi IsozakiMaomi UenoPublished in: ECML/PKDD (1) (2009)
Keyphrases
- free energy
- learning bayesian networks
- statistical data
- bayesian networks
- incomplete data
- belief propagation
- fixed point
- conditional independence
- competitive learning
- upper bound
- structure learning
- posterior distribution
- approximate inference
- data sets
- graphical models
- probability distribution
- machine learning
- markov random field
- least squares
- probabilistic model
- dynamic programming
- pairwise
- metadata