XPost: comp.ai.genetic, comp.ai.neural-nets   
   From: heath@alumni.brown.edu   
      
   mick_from-newsg1@yahoo.com (Mick) wrote in message   
   news:<408e3aa5$1@news.unimelb.edu.au>...   
   > Newbie question here:   
   >   
   >  Does anyone know which systems/topology would be best to   
   >  implement a "square root" function, either using   
   >  neural networks, genetic algorithms, etc   
      
   MLP might be a better NN to use than an RBF.   
      
   If your training data lies in the interval [xmin,xmax]   
   performance should degrade more slowly for x outside the   
   interval with a MLP.   
      
   >  Idea here is 1 input (source value),   
   >   
   >  two outputs:   
   >  1. square root value   
   >  2. Indicator - when set high, calculation has finished.   
      
   Once the net is trained, the output will be more or less   
   instantaneous. No iteration is involved in the trained   
   model.   
      
   >  Idea here is that the training set will be a "simple" table   
   >  of inputs (original values) and outputs (square roots).   
   >  [with indicator output to say when calculation finished, or   
   >  1000 iterations, whichever comes first]   
      
   Again, no iteration is involved with a trained NN.   
      
   Hope this helps.   
      
   Greg   
      
   >  Id like to see which systems could automatically deduce   
   >  a newtonian solution (or other) to this problem.   
   >   
   >  Any ideas, even vague, very welcome!   
   >   
   >  Regards,   
   >            M.   
   >   
   >  	mick_from-newsg1@yahoo.com   
   >   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)