thinQmaster Virtual testing partner Adaptive practice exams that find your weak spots and drill them. Every answer explained. From $69
Enroll

thinQmaster

Find your exam

Five questions free, before you decide.

Step 1 · choose a vendor

In one of our camps? Your practice exams come with the tuition — they are on your study page already.

AI & Machine Learning · Intermediate

L1 Regularization

Adds a penalty proportional to the absolute value of weights, pushing some weights to exactly zero (Lasso).

Definition

L1 regularization produces sparse models by eliminating less useful features, which aids interpretability and performs implicit feature selection. Linear regression with L1 is called Lasso. Elastic Net combines L1 and L2.

Exam tip

L1 = sparsity and feature selection; L2 = small but nonzero weights.

Related terms