From: mlh496@gmail.com   
      
   Suppose you are building a graphical model (Bayesian network).  After   
   you have picked a topology and trained the network, you want to revise   
   the network - make minor changes to the topology by possibly adding a   
   new variable, deleting an edge, etc.   
      
   What techniques are typically used to determine if a small change is   
   worthwhile?  I've read some articles that discuss "quality measures";   
   you accept the change if the quality measure increases.  Intuitively,   
   it seems that there should be some way to consider the marginal   
   decrease in entropy or gain in likelihood.   
      
   Could anyone point me in the right direction?   
      
   All the best,   
   -Michael   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)