Lecture
This lecture covers linear models for classification, starting with a recap of the linear model in dimension D. It then delves into binary classification as regression, adding non-linearity with the logistic sigmoid function, and logistic regression. Decision boundaries and support vector machines are discussed, along with the application of multi-class least-square classification and logistic regression. The lecture concludes with a comparison of linear classifiers on datasets like Iris and MNIST, showcasing their accuracy and training/testing times.