Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture by the instructor covers the concept of mapping non-linearly separable data to higher dimensions using SVM. Various examples are discussed, including 1D and 2D classification, polynomial approximation, and the use of SVM with a polynomial kernel. The lecture also delves into polynomial feature expansion, least-squares formulation, regularization, and the implications of noise in data. Applications of SVM in rainfall prediction and stock price forecasting are explored. The importance of careful curve-fitting methods and the interpretation of linear decision boundaries in high-dimensional spaces are highlighted.