Lower Bounds on Regret for Noisy Gaussian Process Bandit Optimization.
Jonathan ScarlettIlija BogunovicVolkan CevherPublished in: COLT (2017)
Keyphrases
- gaussian process
- lower bound
- regret bounds
- gaussian processes
- upper bound
- regression model
- gaussian process regression
- model selection
- bayesian framework
- gaussian process classification
- np hard
- hyperparameters
- semi supervised
- bandit problems
- latent variables
- approximate inference
- objective function
- random sampling
- online learning
- expectation propagation
- sparse approximations
- gaussian process models
- worst case
- multi armed bandit problems
- probabilistic model
- optimal solution
- bayesian inference
- linear regression
- multi armed bandit
- active learning
- similarity measure