XPost: comp.ai.neural-nets, sci.math.num-analysis, sci.math   
   From: diegoandresalvarez@lycos.co.uk   
      
   Hi!   
      
   we all know that neural networks are a very good algorithm for   
   functional approximation. However they are good when we work in low   
   dimensions (i.e. X, the input vector has less than let's say 15   
   elements), because in higher dimensions the computational overhead   
   becomes the training really computational expensive.   
      
   In this sense, there is any technique that allows an approximation to a   
   function in problems with a higher number of dimensions? lets say X \in   
   R^1000.   
      
   Thanks,   
      
   Diego Andres   
      
   [ 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)