Next: Fitness function
Up: Appendix A: Review of
Previous: Mutation
There are many other operators that may be used to improve the chances of
a fast and accurate convergence of genetic algorithms. The ones used in this
study are
- Elitism: This is the operator that promotes, without change, the best
individual or individuals of the population to the next generation. These
individuals are still eligible for recombination with others to produce
the offspring.
- Niching: Also called sharing is the process of sharing genetic
material between closely related individuals Goldberg and Richardson (1987)
- Creep mutation: Similar to standard mutation but only changes that
result in small perturbations of the decoded model parameters (as opposed
to their binary representation) are allowed. This can be considered a
fine-tuning operator.
Next: Fitness function
Up: Appendix A: Review of
Previous: Mutation
Stanford Exploration Project
11/11/2002