Does gamultiobj produce identically independently distributed? data?

Hi

I store after each iteration of the algorithm gamultiobj the current optimal values of the objective functions in an array. When the optimization is complete, will the data in this array be in general identically independently distributed (iid)?

Thank you very much for your help. Thank you for your help

Answer 1

[Walter Roberson]

No. The members of the population that survive are going to be the ones that are more "fit", which is likely to cover only a subset of the available parameter space, so the survivors are going to be biased rather than statistically independent. Mutations and cross-overs create related members, not independent members.

[Alan Weiss]

Additionally, gamultiobj attempts to spread the final points approximately uniformly in function space. So the final Pareto points are far from randomly distributed.

Alan Weiss

MATLAB mathematical toolbox documentation