GP-BART: A novel Bayesian additive regression trees approach using Gaussian processes.
Mateus MaiaKeefe MurphyAndrew C. ParnellPublished in: Comput. Stat. Data Anal. (2024)
Keyphrases
- gaussian processes
- regression trees
- covariance function
- linear models
- regression problems
- hyperparameters
- gaussian process
- regression model
- genetic programming
- multi task
- gaussian process regression
- random forests
- cross validation
- model selection
- decision trees
- multi task learning
- linear model
- bayesian inference
- linear regression
- closed form
- em algorithm
- random sampling
- maximum likelihood
- bayesian framework
- sample size
- prior information
- support vector
- noise level
- conditional random fields
- dynamic bayesian networks
- training set
- machine learning
- input space
- training data
- learning problems
- data sets
- parameter settings