In this video, we will learn about one of the state of the art machine learning model which is known as Elastic Net Regression.

In Lasso and Ridge regression, we penalise the parameters of the model and this helps in achieving feature selection as well because the features which are not so, important their parameters become close to zero. And it solves the problem of overfitting as well.

By combining the above two models that are Lasso and the Ridge we develop another algorithm that is known as the Elastic Net model. Lasso regression and the Ridge regression use L1 and L2 norms respectively to achieve the properties mentioned above.

Elastic Net regression overcomes the limitations of the lasso (least absolute shrinkage and selection operator) method which uses a penalty function in its L1 regularization.

Use cases of Elastic Net Regression:
1) Metric learning
2) Portfolio optimisation
3) Cancer prognosis

Related Articles :  https://www.geeksforgeeks.org/implementation-of-elastic-net-regression-from-scratch/