Quantcast

Genotype.evolve()

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Genotype.evolve()

Michael Chan-3
Hi,

I've been reading the source of Genotype.evolve() and something slightly
odd has occurred to me. It seems to me that cross operators and mutation
operators copy fitness values from a child's parents, and when
BulkFitnessFunction is used, a child's fitness won't be updated until
bulkFunction.evaluate() is called. With that in mind, the natural
selectors after genetic operators only depend on the fitness values of
the original population, which could undervalue fit children and
overvalue less fit children. So, would it be more reasonable to evaluate
fitness before natural selection?

Cheers,

Michael


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
jgap-devl mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-devl
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Genotype.evolve()

Klaus Meffert
Hello Michael,

I see your problem. As with your other request, may I ask you if you could
kindly provide an example? Maybe it is the same example as with StringGene
and the IGeneConstraintChecker.

Best

Klaus

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of
> Michael Chan
> Sent: Sunday, September 25, 2005 8:03 PM
> To: [hidden email]
> Subject: [jgap-devl] Genotype.evolve()
>
> Hi,
>
> I've been reading the source of Genotype.evolve() and
> something slightly odd has occurred to me. It seems to me
> that cross operators and mutation operators copy fitness
> values from a child's parents, and when BulkFitnessFunction
> is used, a child's fitness won't be updated until
> bulkFunction.evaluate() is called. With that in mind, the
> natural selectors after genetic operators only depend on the
> fitness values of the original population, which could
> undervalue fit children and overvalue less fit children. So,
> would it be more reasonable to evaluate fitness before
> natural selection?
>
> Cheers,
>
> Michael




-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
jgap-devl mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jgap-devl
Loading...