Sampling hyperparameters in hierarchical models: Improving on Gibbs for high-dimensional latent fields and large datasets.
Richard NortonJ. Andrés ChristenColin FoxPublished in: Commun. Stat. Simul. Comput. (2018)
Keyphrases
- hyperparameters
- hierarchical models
- random sampling
- parameter space
- high dimensional
- posterior distribution
- markov chain monte carlo
- hierarchical model
- latent variables
- sample size
- maximum a posteriori
- model selection
- bayesian inference
- bayesian framework
- cross validation
- support vector
- closed form
- gaussian process
- probabilistic model
- low dimensional
- active learning
- generative model
- hierarchical structures
- maximum likelihood
- markov random field
- data points
- learning algorithm
- em algorithm
- dimensionality reduction
- incremental learning
- noise level
- prior information
- data mining
- parameter settings
- incomplete data
- exact inference
- feature space
- image reconstruction
- sliding window
- denoising
- semi supervised