Extension of Linear Regression
Also Known as "L1 Regulisation"
Use: Features Selection
Issue in Linear Regression: Cost function can reach 0, meaning it is a 100% accurate model which is never true.
Therfore, Ridge Regression - L2, Lasso Regression - L1 & Elastic Net adjust the cost functions to give better accurate models
Improves on the following factors
Adds a "Ridge Hyper-parameters for adjusting cost function such that the cost function is always >= 0
Similar to Ridge, but Removes, not just reduce the less important features
Useful for feature selection