On the computation of marginal likelihood via MCMC for model selection and hypothesis testing.
Fernando LlorenteLuca MartinoDavid Delgado-GómezJavier López-SantiagoPublished in: EUSIPCO (2020)
Keyphrases
- marginal likelihood
- model selection
- hypothesis testing
- hypothesis tests
- hyperparameters
- cross validation
- parameter estimation
- gaussian process
- information criterion
- approximate inference
- expectation propagation
- model selection criteria
- posterior distribution
- markov chain monte carlo
- sample size
- monte carlo
- mixture model
- variable selection
- markov chain
- regression model
- feature selection
- closed form
- generalization error
- variational bayes
- bayesian methods
- selection criterion
- machine learning
- confidence intervals
- statistical tests
- bayesian information criterion
- graphical models
- upper bound
- bayesian inference
- probabilistic inference