Gaussoids are two-antecedental approximations of Gaussian conditional independence structures.
Tobias BoegePublished in: Ann. Math. Artif. Intell. (2022)
Keyphrases
- conditional independence
- gaussian model
- bayesian networks
- random variables
- graphical models
- probability distribution
- structure learning
- causal models
- learning bayesian networks
- directed acyclic graph
- axiomatic characterization
- maximum likelihood
- possibility theory
- statistical independence
- bayesian network structure
- gaussian distribution
- conditional independencies
- causal independence
- causal discovery
- chain graphs
- structure and parameter learning
- closed form
- probabilistic model
- marginal distributions
- probability theory
- markov property
- markov networks
- gaussian mixture model
- image segmentation
- conditional independence tests
- social networks