Ted,   
      
   Thank you for the reply!   
      
   Do you think my scheme for avoiding the combinatorial explosion (my   
   reply to the first post) has merit?   
      
   Lets say you have a number of actions, where each action sends a   
   signal to   
   a different circuit.   
      
   (back to the conditions laid out in my reply)   
      
   Action 5 will toggle the second bit of state 9 to give state D, which   
   is closer to state F than   
   state 9.   
      
   Action 10 will toggle the third bit of state D to the desired state F.   
      
   If the plan fails, do another cycle of learning.   
      
   In real life, this could be considered an obvious solution, where   
   changing state   
   outside of 1XX1 would be a surprising solution.   
      
   [ 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)