On the information-theoretic limits of graphical model selection for Gaussian time series.
Gabor HannakAlexander JungNorbert GoertzPublished in: EUSIPCO (2014)
Keyphrases
- information theoretic
- model selection
- mutual information
- cross validation
- information theory
- hyperparameters
- sample size
- parameter estimation
- feature selection
- information bottleneck
- regression model
- information theoretic measures
- mixture model
- machine learning
- generalization error
- marginal likelihood
- selection criterion
- information criterion
- model selection criteria
- statistical inference
- marginal distributions
- log likelihood
- variable selection
- gaussian distribution
- maximum likelihood
- kernel matrix
- gaussian process
- learning problems
- kernel methods