[galib] steady state documentation contradiction
Eyal
eyaler at yahoo.com
Tue May 23 17:26:57 EDT 2006
carefully going over the code and running some tests,
I am now quite sure that the former behavior described
in the API is correct, while the in-code documentation
is obsolete.
to sum up: the Steady State behavior is completely
elitistic - better individuals never being lost.
--- Eyal <eyaler at yahoo.com> wrote:
> The API documentation states that the steady state
> algorithm favors high score members of the previous
> generation over low-score offsprings: "Each
> generation
> the algorithm creates a temporary population of
> individuals, adds these to the previous population,
> then removes the worst individuals in order to
> return
> the population to its original size." and "Newly
> generated offspring are added to the population,
> then
> the worst individuals are destroyed (so the new
> offspring may or may not make it into the
> population,
> depending on whether they are better than the worst
> in
> the population)."
> On the other hand the code documentation in
> GASteadyStateGA.C states the opposite, i.e. that
> low-score offsprings are preffered over high score
> members of the previous generation: "we pick from
> the
> current population and replace its members with the
> new ones we create. We replace the worst members of
> the preceeding population. If a genome in the tmp
> population is worse than one in the main population,
> the genome in the main population will be replaced
> regardless of its better score." and "Replace the
> worst genomes in the main population with all of the
> individuals we just created."
> Which is correct?
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam
> protection around
> http://mail.yahoo.com
> _______________________________________________
> galib mailing list
> galib at mit.edu
> http://mailman.mit.edu/mailman/listinfo/galib
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
More information about the galib
mailing list