Tighter Bounds on the Log Marginal Likelihood of Gaussian Process Regression Using Conjugate Gradients.
Artem ArtemevDavid R. BurtMark van der WilkPublished in: CoRR (2021)
Keyphrases
- gaussian process regression
- marginal likelihood
- gaussian process
- upper bound
- lower bound
- gaussian processes
- model selection
- regression model
- hyperparameters
- approximate inference
- expectation propagation
- bayesian framework
- latent variables
- semi supervised
- cross validation
- worst case
- random sampling
- free energy
- information criterion
- learning algorithm
- incremental learning
- closed form
- em algorithm
- prior knowledge
- clustering algorithm