Linear Models: Classification Basics

This lecture covers the fundamentals of linear models for classification, including hyperplanes, multi-output prediction, non-linearity, and probabilistic interpretation. It also delves into logistic regression training, dealing with multiple classes, and decision boundaries in detail. The lecture concludes with an overview of support vector machines, discussing maximum margin classifiers, slack variables, and the SVM formulation. Various properties and examples of k-Nearest Neighbors are explored, highlighting its applications in classification and regression tasks, along with the curse of dimensionality and strategies to mitigate it.