Gaussoids are two-antecedental approximations of Gaussian conditional independence structures.
Tobias BoegePublished in: CoRR (2020)
Keyphrases
- conditional independence
- gaussian model
- bayesian networks
- graphical models
- learning bayesian networks
- random variables
- probability distribution
- structure learning
- causal models
- directed acyclic graph
- possibility theory
- bayesian network structure
- probability theory
- statistical independence
- axiomatic characterization
- structure and parameter learning
- causal independence
- maximum likelihood
- marginal distributions
- closed form
- higher order
- decision trees
- conditional independence tests
- social networks