Gaussian Process Uniform Error Bounds with Unknown Hyperparameters for Safety-Critical Applications.
Alexandre CaponeArmin LedererSandra HirchePublished in: CoRR (2021)
Keyphrases
- error bounds
- gaussian process
- hyperparameters
- safety critical
- model selection
- cross validation
- bayesian framework
- bayesian inference
- gaussian processes
- closed form
- prior information
- support vector
- formal methods
- random sampling
- em algorithm
- theoretical analysis
- maximum likelihood
- noise level
- maximum a posteriori
- gaussian process regression
- sample size
- worst case
- posterior distribution
- covariance function
- incremental learning
- marginal likelihood
- variational bayes
- incomplete data
- regression model
- parameter settings
- expectation propagation
- latent variables
- missing data
- missing values
- expectation maximization
- supervised learning
- data sets