Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
This lecture covers Ridge Regression, a method to handle multicollinearity in linear models by adding a 'ridge' to the design matrix. It standardizes the design matrix and replaces ZT Z by ZTZ+AI to stabilize inversion. The lecture also discusses the shrinkage viewpoint of Ridge Regression, the bias and variance tradeoff, and the LASSO method as a relaxation of best subsets selection.