Genetic Algorithm and Direct Search Toolbox

Setting Options for the Genetic Algorithm

To get the best results from the genetic algorithm, you usually need to experiment with different options. Selecting the best options for a problem involves trial and error. This section describes some ways you can change options to improve results. For a complete description of the available options, see Genetic Algorithm Options.

This section covers the following topics:

• Diversity
• Population Options
• Fitness Scaling Options
• Selection Options
• Reproduction Options
• Mutation and Crossover
• Mutation Options
• The Crossover Fraction
• Example -- Global Versus Local Minima
• Setting the Maximum Number of Generations
• Using a Hybrid Function
• Vectorize Option

Running ga from an M-File

Diversity

© 1994-2009 The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS