Naive imputation implicitly regularizes high-dimensional linear models.

Alexis Ayme Claire Boyer Aymeric Dieuleveut Erwan Scornet

Published in: ICML (2023)

Keyphrases

linear models
high dimensional
variable selection
additive models
missing values
linear model
missing data
low dimensional
high dimensional data
gaussian processes
nonlinear models
dimension reduction
data points
high dimensionality
linear regression
nearest neighbor
parameter space
feature space
cross validation