GP-BART: a novel Bayesian additive regression trees approach using Gaussian processes.
Mateus MaiaKeefe MurphyAndrew C. ParnellPublished in: CoRR (2022)
Keyphrases
- gaussian processes
- regression trees
- covariance function
- linear models
- regression problems
- hyperparameters
- gaussian process
- genetic programming
- multi task
- regression model
- gaussian process regression
- cross validation
- multi task learning
- random forests
- linear model
- bayesian inference
- linear regression
- model selection
- decision trees
- bayesian framework
- conditional random fields
- maximum likelihood
- support vector
- incremental learning
- random sampling
- closed form
- dynamic bayesian networks
- approximate inference
- probabilistic model
- noise level
- prior information
- graphical models
- least squares
- semi supervised
- sample size
- genetic algorithm