next up previous

4.3.7 Algorithm Performance     continued...

Figure 25 shows the progress in reducing the objective function for the same search. Both the objective function of the best individual within each population and the population average objective are shown (note that the scales are different). These are the two standard measures of progress in an ES run. The difference between these two measures is indicative of the degree of convergence in the population.

Figure 25: Minimization of the Two-Dimensional Rosenbrock Function by an Evolution Strategy: Objective Reduction.

The code used in this example was Schwefel's KORR program. This and other ES programs are available (on diskette) with his book Evolution and Optimum Searching [69] in both FORTRAN and C. The book also contains fully commented listings of the FORTRAN source codes.