Information-theoretic limits of graphical model selection in high dimensions.
Narayana P. SanthanamMartin J. WainwrightPublished in: ISIT (2008)
Keyphrases
- information theoretic
- model selection
- high dimensions
- mutual information
- cross validation
- information theory
- high dimensional data
- high dimensional
- hyperparameters
- feature selection
- sample size
- parameter estimation
- machine learning
- mixture model
- poor quality
- statistical inference
- log likelihood
- high dimensional spaces
- gaussian process
- generalization error
- selection criterion
- model selection criteria
- variable selection
- dimensionality reduction
- computer vision
- learning algorithm
- euclidean distance
- similarity search
- feature vectors
- image segmentation