home   mini-CV   news   research   publications   teaching   PhD students   student projects   talks/tutorials   code   memberships

Michael Biehl: Teaching

Courses regularly given in the University of Groningen Computer Science programme.
For external lectures and tutorial see talks/tutorials.. Supervised BSc, MSc theses and internship projects can be found at here..
Course descriptions and timetables (links given below) are available for University of Groningen students and staff.
Tentative schedule for current and forthcoming teaching periods

Introduction to Machine Learning (BSc level, shared teaching) (previously: Introduction to Intelligent Systems)
Course description    timetable
The purpose of this course is to introduce the concepts of machine learning, including unsupervised and supervised
learning. Unsupervised learning includes dimensionality reduction, clustering, and density-based estimation methods.
Supervised learning is examplified in terms of regression and classification problems. Furthermore, evaluation and
validation methods and the problem of over-fitting are discussed. Models and learning algorithms are illustrated
in terms of selected practical applications.

Modelling and Simulation (MSc level, shared teaching)
Course description    timetable
Modelling and simulation techniques play an increasingly important role in a variety of scientific disciplines.
This course covers some of the most important techniques used in the simulation of systems in physics, chemistry,
biology, sociology or other fields. Examples include models of deterministic chaos, growth processes, game theory
and neural networks. The emphasis, however, is be on the basic concepts and the simulation techniques,
rather than on the actual applications. Numerical methods used in the courseinclude discrete iterations, integration of
differential equations, Monte Carlo simulations and more.

Neural Networks and Compuational Intelligence (MSc level)
Course description    timetable
This module provides an introduction to neural networks and related concepts in machine learning. We discuss
different types of network architectures and their usefulness and limitations in classification or regression problems.
In this context, the corresponding training algorithms is in the focus. Besides their practical implementation, we will address
theoretical aspects, e.g. with respect to their convergence behaviour. The list of topics includes: perceptron training,
support vector machines, gradient-based training, testing and validation methods, multilayered neural networks, shallow
and deep networks as well asalternative architectures.