CAP 5512 Evolutionary Computation

Summary

• To discuss the field of evolutionary computation, focusing on theory and applications of genetic algorithms.

  This course will provide students with the background and knowledge to implement EC algorithms, discuss trade-offs between different variations of EC algorithms, and discuss issues related to the application and performance evaluation of EC algorithms.

• To give students a better understanding of the academic research process.

  Students will be asked to read, review, present, and discuss papers from scholarly journals. Students will also be involved in several projects, including a final project which must be proposed, implemented, written up as a paper, and presented to the class.

Course Structure

• A major component of this course involves reading papers from journals and conferences on Evolutionary Computation. You will be asked to write a one to two page critique of all papers discussed in class. You will be asked to give one or two presentations to review an assigned paper, discussing the strengths and weaknesses of the paper. You will be graded on your understanding of the paper content, on your review of the paper, and on your presentation of the material. In addition, there will be an exam during the semester to test comprehension of the assigned papers.

• There will be one or two programming assignments to introduce you to evolutionary computation.

• Each student will also work on a final research project on a related topic of interest to you. You will be asked to write a short proposal of the work you wish to pursue, review background work, implement the project, write a paper describing your efforts and results, and present the paper in class.

• This is a graduate research oriented course. We want to hear your opinions and comments. Part of your grade will be based on class participation. Participation includes, but is not limited to, class attendance, asking questions, and participation in discussion.

Example Projects

• Britt Crawford and Annie S. Wu (2003). Regulating population size with self-organized criticality. In the Proceedings of the Genetic and Evolutionary Computation Conference (GECCO) 2003 Late Breaking Papers .

• Ozlem O. Garibay, Ivan I. Garibay, and Annie S. Wu (2003). The modular genetic algorithm: Exploiting regularities in the problem space. To appear in the Proceedings of the International Symposium on Computer and Information Systems.

• Shiyuan Jin, Ming Zhou, and Annie S. Wu (2003). Sensor network optimization using a genetic algorithm. In the Proceedings of the 7th World Multiconference on Systemics, Cybernetics, and Informatics , Orlando, FL. [pdf]

• Robert M. Patton, Annie S. Wu, and Gwendolyn H. Walton (2003). A genetic algorithm approach to focused software usage testing. Accepted for publication in Annals of Software Engineering. [pdf] [Additional info]

• Viet C. Trinh, Gregory A. Holifield, and Annie S. Wu (2003). Applied cloning techniques for a genetic algorithm used in evolvable hardware design. In the Proceedings of the 7th World Multiconference on Systemics, Cybernetics, and Informatics , Orlando, FL. [pdf]

• Han Yu, Ning Jiang, and Annie S. Wu (2003). Simulating GA search in a dynamic grid environment. In the Proceedings of the GECCO 2003 Late Breaking Papers .

• Gregory A. Holifield and Annie S. Wu (2002). Population genetics of regulatory genes: A genetic algorithm model. In the Proceedings of the GECCO 2002 Late Breaking Papers .

• Hal Stringer and Annie S. Wu (2002). A simple method for detecting domino convergence and identifying salient genes within a genetic algorithm. In Proceedings of the 2002 GECCO , New York, NY, 594-601. [ps] [ps.gz]

• Ayse S. Yilmaz and Annie S. Wu (2002). The effect of diploidy on integer representations. In the Proceedings of the GECCO 2002 Late Breaking Papers .