Artificial Neural Networks

This lecture presents standard algorithms and new developments of  feedforward Artificial Neural Networks, their functioning, application domains, and connections to more conventional mathematical methods. Examples show the potential and limitations of the methods. Supervised as well as unsupervised learning methods are introduced.

In detail:
1) Introduction, some biological facts
2) Mathematical foundations: probability theory and partial derivatives
3) One layer networks and linear discriminants
4) Multilayer networks and error backpropagation
5) Universality of two-layer networks
6) Radial basis function networks
7) Neuronal maps: Kohonen network, Growing Neural Gas
8) Optimization methods

The course will be given in English upon request.

Grades and credits are given according to the percentage of solved problems in exercise 310012 and presentation of a solution during the exercise.

Lecturers

Details

Course type
Lectures
Credits
5 CP
Term
Winter Term 2017/2018

Dates

Lecture
Takes place every week on Friday from 12:15 to 14:00 in room HZO 100.
First appointment is on 13.10.2017
Last appointment is on 02.02.2018
Exercise
Takes place every week on Wednesday from 14:00 to 15:00 in room ND 03/99.
First appointment is on 18.10.2017
Last appointment is on 31.01.2018
Exercise
Takes place every week on Wednesday from 15:00 to 16:00 in room ND 03/99.
First appointment is on 18.10.2017
Last appointment is on 31.01.2018
Exercise
Takes place every week on Wednesday from 16:00 to 17:00 in room ND 03/99.
First appointment is on 18.10.2017
Last appointment is on 31.01.2018
Exercise
Takes place every week on Wednesday from 17:00 to 18:00 in room ND 03/99.
First appointment is on 18.10.2017
Last appointment is on 31.01.2018

Certificate: Upon successful completion of the exercises (1 HPW)

Literature:

  • C. M. Bishop, Neural Networks for Pattern Recognition, 1995 Clarendon Press, Oxford.
  • S. Haykin, Neural Networks and Learning Machines, 3rd edition, 2003, Pearson, New Jersey

The Institut für Neuroinformatik (INI) is a central research unit of the Ruhr-Universität Bochum. We aim to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory systems and while acting in those environments through effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental approaches from psychology and neurophysiology as well as theoretical approaches from physics, mathematics, electrical engineering and applied computer science.

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210