Classification with GMM and kNN

This lecture covers the classification process using Gaussian Mixture Models (GMM) and k-Nearest Neighbors (kNN). The instructor explains how to model classes with GMM and determine boundaries using ML discriminant rules. The lecture also delves into the Curse of Dimensionality, computational costs, and the comparison results of different models. Practical exercises involve finding boundaries with GMM, understanding the impact of covariance matrices, and analyzing classification errors. The importance of cross-validation and the choice of training/testing ratios are highlighted, along with the application of kNN for classification. The lecture concludes with a demonstration of drawing boundaries with different k values in kNN.