|
gptkbp:instanceOf
|
gptkb:model
statistical analysis
regression analysis method
|
|
gptkbp:advantage
|
automatic variable selection
prevents overfitting
|
|
gptkbp:canSetCoefficientsToZero
|
true
|
|
gptkbp:category
|
supervised learning
sparse modeling
|
|
gptkbp:field
|
gptkb:machine_learning
data science
statistics
|
|
gptkbp:form
|
minimize (1/2n)||y - Xβ||^2_2 + λ||β||_1
|
|
gptkbp:fullName
|
gptkb:Least_Absolute_Shrinkage_and_Selection_Operator
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
LASSO regression
|
|
gptkbp:hyperparameter
|
lambda
|
|
gptkbp:implementedIn
|
gptkb:MATLAB
gptkb:scikit-learn
R
|
|
gptkbp:introduced
|
gptkb:Robert_Tibshirani
|
|
gptkbp:introducedIn
|
1996
|
|
gptkbp:limitation
|
can select at most n variables if n < p
can be unstable with highly correlated variables
|
|
gptkbp:objective
|
minimize sum of squared errors plus lambda times sum of absolute values of coefficients
|
|
gptkbp:penalty
|
L1 norm
|
|
gptkbp:relatedTo
|
gptkb:Elastic_Net
gptkb:Ridge_regression
|
|
gptkbp:shrinksCoefficients
|
true
|
|
gptkbp:usedFor
|
feature selection
linear regression
regularization
|
|
gptkbp:bfsParent
|
gptkb:Ordinary_least_squares
gptkb:Ordinary_Least_Squares
|
|
gptkbp:bfsLayer
|
7
|