Joined: Jan. 2006
|Quote (Cubist @ Mar. 04 2009,03:10)|
|Quote (Richardthughes @ Mar. 02 2009,18:10)|
|variance to the actual x and y values given. cull the bottom z%, mutate back up to full population, rinse, repeat.|
So the "fitness" of each "entity" is simply the numerical value of the expression? Hmmm... if I'm not mistaken, that basically works out to "highest value wins", and I'm not sure how useful/informative/exciting that would be. What about something more like "absolute value of (X - FitnessValue)" for the fitness function? You could get a wider range of results, depending on whether X is an integer, or rational, or irrational, or whatever. Or I think so, anyway...
No fitness would be either the MAD, MSE or MAPE* of the variance between the actual equation and the 'organism' equation for known X's. Low numbers being better.
*Mean Absolute Deviation, Mean Squared Error, Mean Absolute Percentage Error - all should work equally well in this case.
"Richardthughes, you magnificent bastard, I stand in awe of you..." : Arden Chatfield
"You magnificent bastard! " : Louis
"ATBC poster child", "I have to agree with Rich.." : DaveTard
"I bow to your superior skills" : deadman_932
"...it was Richardthughes making me lie in bed.." : Kristine