Tuesday, August 15, 2006

Segregative Genetic Algorithms (SEGA): A Hybrid Superstructure

This sounds like a robust and efficient solution for the issue of premature convergence (the child is no better than the parent) in genetic algorithms.
http://www.heuristiclab.com/publications/papers/affenzeller01d.pdf

1 comment:

Stefan Wagner said...

Please note that there is a newer and even more powerful algorithm called "Self-Adaptive Segregative Genetic Algorithm with Simulated Annealing Aspects" (SASEGASA). Based on the ideas of SEGA, self-adaptive selection pressure steering and offspring selection were additionally introduced in SASEGASA. By this means, SASEGASA shows an even better performance in terms of retarding premature convergence and achieving solutions of very high quality.

More information about SASEGASA can be found on the publications page of Michael Affenzeller:

http://www.heuristiclab.com/publications/affenzeller.html

Business Analytics

Business Analytics

About Me

My photo
See my resume at: https://docs.google.com/document/d/1-IonTpDtAgZyp3Pz5GqTJ5NjY0PhvCfJsYAfL1rX8KU/edit?hl=en_USid=1gr_s5GAMafHRjwGbDG_sTWpsl3zybGrvu12il5lRaEw