L1 (Lasso) vs. L2 (Ridge) regularization

Machine Learning Medium Seen in real interview

What is L1 and L2 regularization? What are the differences between the two?

L1 is Lasso and it penalizes the sum absolute values of the coefficients.

L2 is ridge, and it penalizes the sum of squared coefficients.

L1 (LASSO)	L2 (RIDGE)
Differentiable exept at zero	Easily differentiable
Will zero out coefficients that don’t contribute	Will penalize greater coefficients more
Used for feature selection	Used for feature regularization

Topics

L1, L2, Redularization, Lasso, Ridge

How did you find the solution? Correct Needs more details Incorrect Did not read

Did you find this question to be: Easy Medium Hard

Overall, do you feel this is a good interview question?

Have you seen this question in an interview? If yes, with what company?