Beefy Boxes and Bandwidth Generously Provided by pair Networks vroom
We don't bite newbies here... much
 
PerlMonks  

Re: Genetic Programming or breeding Perls

by t'mo (Pilgrim)
on May 03, 2001 at 00:22 UTC ( #77499=note: print w/ replies, xml ) Need Help??


in reply to Genetic Programming or breeding Perls

I realize that this thread may not be followed anymore, but if it does (and especially if gumpu is out there listening), it seems to me that each Individual ought to know its own fitness. In other words, FITNESS should be a field in each Individual, and FITNESSES should not be field in Population.

Any thoughts?


Comment on Re: Genetic Programming or breeding Perls
Select or Download Code
Re: Re: Genetic Programming or breeding Perls
by gumpu (Friar) on May 05, 2001 at 12:27 UTC

    You are right. FITNESS should be a field in each Individual. Fitness is something associated with an individual. Population should only keep track of the statistics of all the individuals in the population. Bad design choice; probably had a case of 'premature optimization' :)

    Have Fun

Re: Genetic Programming or breeding Perls
by kwilliams (Sexton) on May 24, 2001 at 19:22 UTC
    I disagree. Fitness is a combination of the individual abilities and the constraints of the environment. If you put an individual in a different environment its fitness for that environment will change.

      Yes, fitness is related to the environment. But...

      In this case, if the algorighm becomes simpler if FITNESS is more tightly associated with the INDIVIDUAL. For example, this:

      sub choose { my $self = shift; my $f = rand(1.0); my $index = 0; my $sum = 0.0; foreach my $fitnes (@{$self->{FITNESSES}}) { $sum += $fitnes; return ${$self->{INDIVIDUALS}}[$index] if $sum >= $f; ++$index; } die "can't select an individual"; }

      becomes:

      sub choose { my $self = shift; my $f = rand(1.0); my $sum = 0.0; foreach my $individual (@{$self->{INDIVIDUALS}}) { $sum += $individual->{FITNESS}; return $individual if $sum >= $f; } die "can't select an individual"; }

      which I consider an improvement, not only due to the fact that there's (a little) less code, but the concept of choosing an individual and not a fitness is emphasized.

      Finally, I must admit that I have a bias in relation to the idea you presented. Yes, environment does determine fitness. However, what if you're trying to evolved generalized behavior, i.e., a program that will perform well in any environment, and not simply the one's it was trained in? The little bit of work I've done with GP has been focused in the direction of trying to avoid such "over-training" or specialization.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://77499]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others imbibing at the Monastery: (9)
As of 2014-04-18 07:23 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    April first is:







    Results (463 votes), past polls