 |
|
| Computers Forum Index » Computer Artificial Intelligence - Genetic » Genetic Programming Representations... |
|
Page 1 of 1 |
|
| Author |
Message |
| Ken T.... |
 Posted: Thu Nov 20, 2008 5:50 am |
|
|
|
Guest
|
On Wed, 19 Nov 2008 22:18:08 +0000, I wrote:
Quote: I took my first foray into genetic programming last weekend. I'm
familiar with other kinds of evolutionary computation, but I've never
before tried to evolve programs.
.....
One more thing I wanted to ask about. I implemented intron counting
based on what I read in the text I have on Genetic programming. I
thought it would be interesting to see the explosive intron growth
mentioned in the book.
Well, my results were somewhat different than that reported in the text.
Introns rapidly take over more than 90% of the program. If I stop at
this point, I'm still missing a good possibility of improvement. If I
continue up to 99%, there is still no assurance that my run has
stagnated.
The only think I can think of that might be happening is that some
individuals in the population have been destroyed so utterly that most of
their length isn't being executed. There must be many of these because a
sample of 5% of the population results in > 95% introns.
Maybe I should just ignore this metric? Any hints on what I might be
doing wrong?
--
Ken T.
When you choose the lesser of two evils, always remember that it is
still an evil.
-- Max Lerner |
|
|
| Back to top |
|
|
|
|
|
All times are GMT
The time now is Thu Mar 11, 2010 10:01 pm
|
|