Data Analysis and Manifold Learning

Agenda: The course will start in January and will finish in April. There will be approximatively 2-3 lectures per month. Who should attend: PhD candidates and Master students who are interested in applying machine learning methodologies to their research. The course is open to any researcher interested in manifold learning tools for data analysis. 1 Course Objective The objective of this course is to familiarize PhD candidates and Master students in computer science, electrical engineering, and applied mathematics with data analysis methodologies that fall within the topic of statistical machine learning, (Hastie et al., 2009) and which have their roots in algebraic graph theory and heat-diffusion in Riemannian geometry. Within the light of these powerful mathematical tools, the data will be analysed from the perspective of their intrinsic geometry. There are many fields that can profit from these methods, such as: • Visual computing (images, image collections, and videos) • Graphical and animation methods (voxels, meshes, 3D scans)