Lecture
This lecture provides a recap on clustering, semi-supervised clustering, and classification, explaining the concepts of clustering and classification without labels, the use of classifiers like neural networks, decision trees, and support vector machines. It delves into the history and applications of Support Vector Machine (SVM), detailing its principles, optimal classifier learning, and the determination of the optimal separating hyperplane. The lecture also covers non-linear classification, the effect of the width of the Gaussian kernel, and the hyperparameters of SVM. It concludes with an exercise on SVM, discussing support vectors and the shape of the separating hyperplane for different kernels.