Advances in Evolutionary Computation
CS 6320, Spring 2008, G01, TR 8:20PM-9:35PM, 417 CLARK HALL

Instructor:
Martin Pelikan
320 Computer Center Building (CCB)
E-mail: pelikan@cs.umsl.edu
WWW: http://www.cs.umsl.edu/~pelikan/
Note: Do not call, best contact via email.

Office hours
TBA, 320 CCB
or by appointment (send email to arrange)

Prerequisites
CS 5320 or consent of instructor

Textbook

• David E. Goldberg (2002). The Design of Innovation: Lessons from and for competent genetic algorithms. Springer, 1st edition.
• Material not covered in the textbook will be provided in the form of printouts and web links.

Grading Policy

• Essays...25%
• Presentations...25%
• Final project...50%

Covered topics

• Multiobjective genetic algorithms.
• Niching.
• Linkage learning.
• Estimation of distribution algorithms.
• LINC, LIMD, and GEMGA.
• Efficiency enhancement.
• Parallel genetic algorithms.
• Evaluation relaxation.
• Genetic programming.

Presentations and essays (not related to the projects)

• Each week or so we will cover one topic in EC.
• For each topic one student will read several papers/chapters, and lead a lecture on this topic (give presentation, lead discussion, and answer questions).
• All other students (non-presenters) will read at least one of those assigned papers/chapters (selection is decided by the instructor) and write a 1-page (not longer) essay on the topic (before the corresponding lecture).
• Every student is required to lead at least one lecture.

Final projects

• Each student must complete a final project (individually).
• Each project will result in a paper of up to 10 pages (11-point or 12-point sized font).
• Tentative topics for projects are listed here and more can be presented upon request.
• Each student must give two 15-30 minute in-class presentations about the project. The first presentation will cover the goals and preliminary results of the project, the second one will cover the final results of the project.

Additional info

• If you don't have all the prerequisities, you MUST contact me ASAP.
• Covered topics might evolve over the course of the semester based on progress and interest.
• Any extensions only with prior consent of the instructor and only under extraordinary circumstances (at instructor's discretion). If your extension is not granted prior to the deadline, you may get 0.
• Grades will be curved at the discretion of the instructor.
• No delayed grade will be given unless really special circumstances are proven.
• Maintaining confidentiality of student grades: Click here.
• Student conduct code including the campus policy on academic dishonesty: Click here.
• Cancellation of student registration: Click here.
• Wait list: Click here.

Short bio of the instructor:
Martin Pelikan received Ph.D. from the Dept. of Computer Science at the University of Illinois at Urbana-Champaign in 2002. He joined the Dept. of Math and Computer Science at the University of Missouri at St. Louis in August, 2003. Currently, he is an assistant professor of computer science. Pelikan's research focuses on genetic and evolutionary computation. He worked at the Slovak University of Technology at Bratislava, the German National Center for Information Technology at Sankt Augustin, the Illinois Genetic Algorithms Laboratory (IlliGAL) at the University of Illinois at Urbana-Champaign, and the Swiss Federal Institute of Technology (ETH) at Zurich. Pelikan's most important contributions to genetic and evolutionary computation are the Bayesian optimization algorithm (BOA), the hierarchical BOA (hBOA), and the scalability theory for BOA and hBOA. BOA and hBOA combine machine learning with genetic and evolutionary algorithms to create optimizers that can solve broad classes of optimization problems in a robust and scalable manner with few or no parameters. BOA and hBOA are among the most advanced and powerful genetic and evolutionary algorithms.