Fast variable selection makes scalable Gaussian process BSS-ANOVA a speedy and accurate choice for tabular and time series regression.
David S. MebaneKyle HayesAli BaheriPublished in: CoRR (2022)
Keyphrases
- variable selection
- gaussian process
- model selection
- gaussian processes
- regression model
- cross validation
- hyperparameters
- sample size
- gaussian process regression
- covariance function
- approximate inference
- sparse approximations
- bayesian framework
- machine learning
- unsupervised learning
- high dimensional
- feature selection
- upper bound
- sparse representation
- dimension reduction
- latent variables
- high dimensional data
- posterior distribution
- semi supervised
- data sets