X. Geng (USA)
Genetic Algorithms, Model Identification, Asynchronous Finite State Machines
A model identification problem is solved for Asynchronous Finite State Machines based on Genetic Algorithms. Particularly, different coding schemes and reproduction schemes are studied by analyzing their properties and corresponding system performances. It is shown that problem-specific evolutionary schemes are more effective than a universal solution for genetic algorithms. The effectiveness is judged based on percentage of invalid chromosomes examined during the evolution process and the convergence rate of the algorithms. The evaluation function design and initial population generation are also addressed.
Important Links:
Go Back
