|
gptkbp:instanceOf
|
regression analysis method
|
|
gptkbp:alternativeName
|
L1 regularization
|
|
gptkbp:contrastsWith
|
Ridge regression (L2 regularization)
|
|
gptkbp:doi
|
10.1111/j.2517-6161.1996.tb02080.x
|
|
gptkbp:effect
|
performs feature selection
shrinks some coefficients to zero
|
|
gptkbp:field
|
gptkb:machine_learning
data science
statistics
|
|
gptkbp:fullName
|
gptkb:Least_Absolute_Shrinkage_and_Selection_Operator
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
LASSO
|
|
gptkbp:hyperparameter
|
regularization parameter (lambda)
|
|
gptkbp:introduced
|
gptkb:Robert_Tibshirani
|
|
gptkbp:introducedIn
|
1996
|
|
gptkbp:limitation
|
can select at most n variables if n < p
biased estimates for large coefficients
|
|
gptkbp:penalty
|
L1 norm
|
|
gptkbp:publishedIn
|
gptkb:Journal_of_the_Royal_Statistical_Society,_Series_B
|
|
gptkbp:reduces
|
sum of squared errors plus L1 penalty
|
|
gptkbp:relatedTo
|
gptkb:Elastic_Net
gptkb:Ridge_regression
|
|
gptkbp:solvedBy
|
gptkb:least_angle_regression
coordinate descent
subgradient methods
|
|
gptkbp:usedFor
|
regression
regularization
variable selection
|
|
gptkbp:bfsParent
|
gptkb:Boruta
|
|
gptkbp:bfsLayer
|
5
|