# A very short introduction to Ridge and Lasso regression

--

Regularization helps avoid overfitting of a machine learning model. Ridge and Lasso can be best explained with MLR (Multiple Linear Regression) in mind.

In PCA, when try to reduce the *number *of features. But in Ridge and Lasso regression, we try reduce the *magnitude *of the coefficients. By shrinking the coefficients, Ridge and Lasso regression reduce the model complexity and multi-collinearity, thus help in avoiding overfitting .That’s why they come under the category of shrinkage methods.