Computational-Intelligence

Old News

Lecture Fuzzy Systems

Winter 2008/09

News

The last slides are online.

The last assignment sheet is online.

Overview

General Information

This page contains information about the lecture "Fuzzy Systems" that is read by Prof. Dr. Rudolf Kruse in the winter 2008/09. It is updated during the course.

Fuzzy set theory is an extension of the classical set theory that can model imprecise and vague expressions of natural language such as big, small, hot, cold, etc. Fuzzy logic allows to formalize rules that contain such expressions of natural language. These rules can be utilized to support decision processes. The lecture "Fuzzy Systems" offers an introduction to both fuzzy set theory and fuzzy logic. Moreover it deals with applications of control engineering, approximate inference and data analysis.

At the end of the lecture an interesting topic in current research is touched. The union of two soft computing methods, i.e. fuzzy systems and support vector machines (SVMs), is motivated and outlined. Therefore the concept of SVMs is introduced.

Note that since "Fuzzy Systems" is a master course, both lecture and exercise will be given in English. There might be German assignment sheets on request.

back to top

Schedule and Rooms

 WeekdayTimeRoomBegin
LectureWednesday09:00-11:00G22A-12022.10.2008
ExerciseThursday11:00-13:00G22A-12023.10.2008
back to top

Lecturers

If you have questions regarding the lecture or exercise, please contact (via e-mail if possible) one of the persons named below.

back to top

Conditions for Certificate ("Schein") and Exam

A new assignment sheet containing written and programming assignments is published on this web page every week. The written assignments must me ticked of at the beginning of every exercise. Ticking off an assignment you agree to be able to explain and present the assignment and your solution proposal (which need not to be completely correct. However, you should be prepared thoroughly in order to solve the assignment).

Programming assignments must be submitted in electronic form to Christian Moewes before 8 o'clock in the morning of the corresponding exercise day. Late assignments will not be accepted. The programs must run properly on the SUN pool machines in the faculty of computer science.

The certificate for this course is issued to students who

  • regularly contribute well in the exercises,
  • tick off at least two third of all written assignments,
  • present at least twice a solution to a written assignment during the exercise,
  • submit at least twice a running implementation of a programming assignment, and
  • finally pass a small colloquium (around 10 minutes) or a written test (if there are more than 20 students) after the course.

Certificate colloquiums will be performed only on February 4 and 6, 2009. Please write to Christian Moewes if you want to get an appointment for one of the colloquiums.

Students who want to finish the course with an exam or a marked certificate must

  • regularly contribute well in the exercises,
  • tick off at least half of all written assignments,
  • present at least twice a solution to a written assignment during the exercise,
  • find an examination appointment with Prof. Kruse,
  • officially announce the exam to the examination office two weeks before the exam, and
  • finally pass an oral exam (around 25 minutes) or a written exam (if there are more than 20 students) after the course.

A list with free appointments for oral exams can be found in the examination office of the FIN.

back to top

Prerequisites

You do not need but should have background knowledge about

  • mathematics (especially algebra and convex optimization theory),
  • computer science (algorithms, data structures, etc.), and
  • machine learning or data mining.
back to top

Slides from the Lecture

The slides from the lecture will be published here incrementally as the course proceeds.

back top top

Assignments

The collection of assignment sheets will be updated weekly at this location. For the most recent one, simply consider only the last of all assignment sheets.

back top top

Additional Material

Feel free to check out the following supplementary material that augment the lecture and exercise.

Below you can find some gnuplot scripts from the exercise. Please download them and rename their extensions from "txt" to "plt". Then they should work. For any comments, remarks, critics, etc. please contact Christian Moewes.

back to top

References

  • about fuzzy systems
    • C. Borgelt, F. Klawonn, R. Kruse, D. Nauck (2003). Neuro-Fuzzy-Systeme (3rd edition). Vieweg, Braunschweig/Wiesbaden, Germany.
    • G.J. Klir and B. Yuan (1995). Fuzzy Sets and Fuzzy Logic - Theory and Applications. Prentice Hall, Upper Saddle River, NJ, USA.
    • R. Kruse, J. Gebhardt, and F. Klawonn (1994). Fuzzy-Systeme (2nd edition). Teubner, Stuttgart, Germany.
    • R. Kruse, J. Gebhardt, and F. Klawonn (1994). Foundations of Fuzzy Systems. Wiley, Chichester, United Kingdom.
    • K. Michels, F. Klawonn, R. Kruse, and A. Nürnberger (2002). Fuzzy-Regelung. Springer-Verlag, Heidelberg, Germany.
  • about machine learning data mining
    • D. Hand and M. Berthold (2002). Intelligent Data Analysis: An Introduction (2nd edition). Springer-Verlag, Berlin, Germany.
    • T. Mitchell (1997). Machine Learning. McGraw Hill, New York, NY, USA.
  • about SVM
    • S. Boyd and L. Vandenberghe (2004). Convex Optimization. Cambridge University Press, New York, NY, USA.
    • B. Schölkopf and A.J. Smola (2002). Learning with Kernels. MIT Press, Cambridge, MA, USA.
    • V. Vapnik (1995). The Nature of Statistical Learning Theory. Springer-Verlag, New York, NY, USA.
back to top

Links

back to top
en lang icon de lang icon Printable View - Recent Changes
Page last modified on January 27, 2009, at 09:20 PM by cmoewes