Announcements

General

Lecturer: Prof. dr. Thomas Back (baeck@liacs.nl)
Teaching assistant: Johannes Kruisselbrink (jkruisse@liacs.nl)

Lectures / werkcolleges

Lectures: Wednesday 11:15-13:00, Snellius WI-405.
Werkcolleges: Thursday 11:15-13:00, Snellius WI-312.

Course description

Evolutionary algorithms are search and optimization algorithms gleaned from the model of organic evolution. Their main components are a population of individuals that undergoes an iterative process of fitness evaluation, variation and selection. The existing approaches to evolutionary computation - including e.g. genetic algorithms, evolution strategies, evolutionary programming, genetic programming, classifier systems - all share the same basic model, but are considerably different in their practical instantiations. In the course, we will give an overview of the main representatives of evolutionary algorithms and explain the algorithms in detail. The main theoretical results about these algorithms as well as practical application examples are discussed. The biological background, basic foundations of optimization theory, and relationships to other fields will complete the course.

Grades

Student NumberPA gradeExam gradeFinal grade
03243298.37.98.0
07314557.15.46.0
04333578.05.76.5
03081616.77.87.5
02061216.88.98.0
04092008.38.88.5
03187606.73.24.5

Submissions practical assignment

Alexander Aleman
Antanas Kaziliunas
Erik Gast
Eyal Halm and Michiel Helvensteijn
Timo de Vries and Renuka Autar

Course regulations

The course consists of:

Exam:
The exam is closed-book and will be based on the lecture-slides.
Exam date: Friday January 18th 2008, from 10:00 to 13:00

Practical assignment:
The practical assignment is based on a real-life challenging problem; the autocorrelation problem. It may be done individually or in pairs. The assignment should be implemented in MATLAB and the format of the report should follow the ACM SIG Proceedings Template.

Problem sets:
The problem sets are not mandatory, but are there to help you to get better understanding of the course material.

Final grade:
The final grade is obtained in the following way:

Final grade = 0.6 * exam_grade + 0.4 * practical_assignment_grade

A successful completion of the course will be rewarded with 6ECTS.

Course material

The course material consists of the slides that are used in the lectures.

Lecture slides
0. Organizational issues
1. Introductory examples
2. Biological background
3. Optimization
4. Evolutionary algorithms: basic concepts
5. Genetic algorithms
6. Evolution strategies
Werkcollege slides
- Practical Assignment introduction
- MATLAB introduction

Recommended literature
Thomas Back - Evolution Strategies, Evolutionary Programming, Genetic Algorithms
Oxford University Press, New York 1996
ISBN10: 0195099710 / ISBN13: 9780195099713

Schedule

From time to time, some lectures may be cancelled, and some of the werkcollege-slots may be used for lectures.

DateTimeRoomLecture / Werkcollege
Sep. 5------
Sep. 6------
Sep. 12------
Sep. 13------
Sep. 1911:15 - 13:00405Lecture
Sep. 2011:15 - 13:00312Werkcollege
Sep. 2611:15 - 13:00405Lecture
Sep. 27------
Oct. 3------
Oct. 4------
Oct. 1011:15 - 13:00405Lecture
Oct. 1111:15 - 13:00312Werkcollege
Oct. 1711:15 - 13:00405Lecture
Oct. 18------
Oct. 2411:15 - 13:00405Lecture
Oct. 2511:15 - 13:00312Werkcollege
Oct. 31------
Nov. 111:15 - 13:00312Lecture
Nov. 7------
Nov. 811:15 - 13:00405Lecture
Nov. 1411:15 - 13:00405Lecture
Nov. 1511:15 - 13:00312Werkcollege
Nov. 2111:15 - 13:00405Lecture
Nov. 22------
Nov. 2811:15 - 13:00405Lecture
Nov. 2911:15 - 13:00312Werkcollege
Dec. 511:15 - 13:00405Lecture
Dec. 6------