Conditional independence testing based on a nearest-neighbor estimator of conditional mutual information.
Jakob RungePublished in: AISTATS (2018)
Keyphrases
- conditional independence
- nearest neighbor
- conditional mutual information
- mutual information
- bayesian networks
- selection criterion
- random variables
- probability distribution
- graphical models
- directed acyclic graph
- learning bayesian networks
- knn
- causal models
- axiomatic characterization
- statistical independence
- structure learning
- possibility theory
- high dimensional data
- least squares
- training set
- maximum likelihood
- probability theory
- high dimensional
- image analysis
- bayesian network structure
- chain graphs
- data points
- image segmentation
- maximum a posteriori
- data sets
- model selection
- conditional independencies
- neural network