Lecture 2.1: Linear models for regression

Linear models are the workhorses of many machine learning applications. We will see how we can regularize (simplify) regression models using Ridge and Lasso, and how these can be trained on large datasets using (stochastic) gradient descent and variants thereof. We'll also see alternative loss functions to deal with outliers and fit support vector machines for regression. Slides and notebooks: https://ml-course.github.io/master/