Memorize to generalize: on the necessity of interpolation in high dimensional linear regression.
Chen ChengJohn DuchiRohith KuditipudiPublished in: COLT (2022)
Keyphrases
- linear regression
- high dimensional
- generalized linear models
- least squares
- low dimensional
- regression methods
- regression problems
- ridge regression
- nonlinear regression
- locally weighted
- linear regression model
- high dimensionality
- feature space
- dimensionality reduction
- linear models
- regression method
- regression trees
- dimension reduction
- data points
- decision trees
- variable selection
- parameter space
- high dimensional data
- kernel density estimators
- kernel regression
- linear predictors
- input space
- loss bounds
- multivariate regression