Lecture :
Dr. Laurenz Wiskott
Exercises :
Henning Sprekeler
The exam is scheduled to last 90 minutes and consists of two parts. The first one will consist of general questions, while the second one will test the mathematical side of the lecture.
You may use a formula collection (one double-sided handwritten A4-page) and a pocket calculator.
For those of you who want to get the 'Schein' we highly recommend taking the test exam, as it will provide you with a feeling for the difficulties of the final exam.
Target group:
The course is designed for students of biology, biophysics, physics, mathematics and computer science after the "Vordiplom".
Prerequisite is a solid background in linear algebra and calculus. The course will be mathematically more demanding than the
parts of the "Computational Neuroscience" courses that were taught by L. Wiskott!
Topics :
The course will focus on topics in supervised and unsupervised learning. Apart from standard techniques from the
neural network paradigm (backpropagation, reinforcement learning, self-organizing maps) more abstract methods
in machine learning (support vector machines, graphical models) will be introduced. The emphasis will lie on the theory and technical
applications.
Criteria for the "Schein":
There will be a final written exam.