Nataraj mentions experiments.  This reminds me that we will need a lot 
of experimental testing to see if the new genetic operator gains us 
anything.  In particular, can it enable waybak to achieve smarter ANNs?  
So we need a few people that like to run software, to do this.  

There are really two aspects to this running of the software that we 
need.  I have not mentioned that we don't know what probabilities to use 
when deciding which genetic operator to use.  Our existing software 
simply does single point crossover 2/3 of the time, randomly, and two 
point crossover 1/3.   But now we need a probability for when to do 
number 3, and also, when to reverse a segment when doing number 3.   
Metavolve can be used to come up with good values for these 
probabilities if I code them into the .ini file.

So, this is the first call for people that like to run software, to 
volunteer for this.   (However, the software is not ready, and prolly 
won't be for a couple of months.)

m

-- 

I'm proud of http://ANNEvolve.sourceforge.net.  If you want to write software, 
or articles, or do testing or research for ANNEvolve, let me know.  


-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

Reply via email to