Hyperparameters Tuning in Regularized System Identification with Nonzero Prior Means.
Håvard Bjørgan BjørkøyDamiano VaragnoloPublished in: ECC (2024)
Keyphrases
- hyperparameters
- maximum a posteriori
- prior information
- model selection
- bayesian methods
- parameter settings
- closed form
- cross validation
- random sampling
- gaussian process
- bayesian framework
- em algorithm
- posterior distribution
- bayesian inference
- support vector
- prior knowledge
- regularization parameter
- gaussian processes
- incremental learning
- noise level
- sample size
- maximum likelihood
- image reconstruction
- machine learning
- parameter optimization
- missing values
- expectation maximization
- markov random field
- active learning
- decision trees