Mazes

• The shape of the state space is also important for the choice of chaining methods in RBSs
• Forward and Backward Chaining
• If there is lots of fan in use Forward Chaining
• Fan in means that lots of initial states lead to one (or a small number of) conclusion
• If you use forward chaining, rule applications will quickly converge to an answer, but backward chaining will consider a lot of unnecessary subgoals
• If there is lots of fan out use Backward Chaining
• Fan out means that there are lots of conclusions.
• Only use the rules that are appropriate to the goal.
• If it is really big, use interleaved chaining intelligently.
• Reduce the number of rules applied by negotiating the search space in a wise way.
• For example, imagine a maze where you start in a room, and each exit leads to a series of doors that gets you out of the maze. For this any path of rules will get you out, so just use forward chaining.
• However, if you need to get to a particular exit you should backward chaining. This way you can avoid all of the unnecessary doors.
• Backtracking