| Computers Forum Index » Computer Artificial Intelligence - Genetic » Genetic algorithm without gene variables... |
|
Page 1 of 1 |
|
| Author |
Message |
| Andy... |
 Posted: Thu Sep 24, 2009 8:44 pm |
|
|
|
Guest
|
I'm new to this field, but it's interested me for several years now.
I have an idea and I was wondering whether anyone else has tried it.
My idea is that instead of a common genetic algorithmic design with a
set of variables which are mutated and a fitness function to choose
survivors, the program itself is contained as a genetic variable and
the code is mutated.
The fitness function would be a challenging environment in which a
beneficial mutation would result in more offspring or less likely
death.
I've elaborated this concept into more detail and would like to
discuss it or read up on other's attempts. |
|
|
| Back to top |
|
|
|
| Erik Max Francis... |
| Posted: Fri Sep 25, 2009 2:07 am |
|
|
|
Guest
|
Andy wrote:
Quote: I'm new to this field, but it's interested me for several years now.
I have an idea and I was wondering whether anyone else has tried it.
My idea is that instead of a common genetic algorithmic design with a
set of variables which are mutated and a fitness function to choose
survivors, the program itself is contained as a genetic variable and
the code is mutated.
Yes, there's a whole (quite popular) subfield of genetic algorithms
which concerns this. It's called "genetic programming."
--
Erik Max Francis && max at (no spam) alcyone.com && http://www.alcyone.com/max/
San Jose, CA, USA && 37 18 N 121 57 W && AIM/Y!M/Skype erikmaxfrancis
Life is something to do when you can't get to sleep.
-- Fran Lebowitz |
|
|
| Back to top |
|
|
|
|
