A municipal compost-bin table has dozens of weak optional columns, and the trainer should be allowed to drive many of those weights to zero. Which regularization produces that sparsity?
Select an answer to reveal the explanation.
Short Explanation
Think of dozens of weak optional compost columns, and a trainer that should be allowed to drive many of those weights to zero. L1 regularization. L2 shrinks weights but tends to leave them nonzero.
Full Explanation
L1 regularization drives many weights to exactly zero and acts as implicit feature selection. L2 shrinks weights but tends to keep them nonzero. Polly and extra epochs are not that penalty.