Regularization: L1, L2 and Elastic Net

Control overfitting by penalising coefficient size — and know exactly what each penalty does

Learn how Ridge (L2), Lasso (L1) and Elastic Net penalties shrink or zero out coefficients, why lasso does feature selection and ridge does not, how to pick lambda, and how the same idea shows up in logistic regression, neural nets and gradient boosting.

Practice questions (5)

• Lasso vs Ridge on Correlated Features

  Intermediate · Free
  View →

• Ridge Closed Form by Hand

  Intermediate
  View →

• Standardisation, Intercept and Pipeline Placement

  Intermediate
  View →

• Regularised Logistic Regression and the C Parameter

  Intermediate
  View →

• Regularisation Beyond Linear Models

  Advanced
  View →

Quizzes (1)