Lecture course in sommer semester 2013:

V5E2 - Selected Topics in Numerical Methods in Science and Technology
Mathematical Methods in Machine Learning

Prof. Dr. Jochen Garcke

Location:	Room 6.020, Wegelerstr. 6
Time:	Friday, 10 -12
Replacement Time:	Thursday, 16 - 18 (e.g. lectures after a free Thursday will be moved to the week after or before)
Exams:	1st Period: Thursday 25th July and Friday 26th July
	2nd Period: To be confirmed, either week of 16th of September, or week of 23th of September.
Office hours:	on appointment, e-mail garckeins.uni-bonn.de

Content of the lecture:

Mathematics plays a significant role in the analysis and further development of many machine learning and data mining algorithms. This lecture will cover some of the problems, describe the mathematical foundation of the algorithm and the numerical schemes which can be applied, and gives examples for applications.

The lecture can cover some of the following topics. The particular focus of the lecture will be adapted to the interest and background of the participants.

• Nonlinear Dimensionality Reduction / Manifold Learning
• Kernel Methods
• High Dimensional Data Analysis

Prerequisites:

The lectures on Numerische Mathematik are recommended.

Selected Literature:

• J. Lee and M. Verleysen, Nonlinear Dimensionality Reduction, 2007
• Jianzhong Wang, Geometric Structure of High-Dimensional Data and Dimensionality Reduction, 2011