Bayesian graphical models, intention-to-treat, and the rubin causal Model.
David MadiganPublished in: AISTATS (1999)
Keyphrases
- graphical models
- causal models
- bayesian networks
- conditional independence
- statistical inference
- belief propagation
- random variables
- causal discovery
- probabilistic model
- directed acyclic graph
- approximate inference
- probabilistic graphical models
- causal relationships
- probabilistic inference
- structure learning
- causal relations
- markov networks
- map inference
- belief networks
- exact inference
- chain graphs
- posterior distribution
- graph structure
- marginal probabilities
- conditional random fields
- gaussian processes
- probability distribution
- maximum likelihood
- factor graphs
- bayesian inference
- statistical relational learning
- posterior probability
- influence diagrams
- conditional probabilities
- loopy belief propagation
- branch and bound
- message passing
- hidden variables