This preview shows page 1. Sign up to view the full content.
Unformatted text preview: to show how the
two subtrees connect. Note that the root node AA is shown in both halves.
B1 The Game Tree (7 points)
You are provided with static values at all the nodes. In this part of the problem, you
are to ignore the static values at intermediate notes. You are to use only the values
on the leaf nodes.
Perform the alpha-beta algorithm on the graph tree on the next two pages.
1) Use the boxes to help you perform your alpha-beta calculations as needed.
2) Draw an 'X' through any branch that is blocked from further consideration by
an alpha-beta calculation.
3) Assume no progressive deepening 3 4 5 B2 Evaluations (20 points)
How many nodes were statically evaluated and which were they?
# Evals __________ Nodes: _________________________________________________ B3 Move (3 points)
What move is chosen as the best? What node in the deepest level is the maximizing player trying to
AA-->__________ 6 Moving Toward: __________ Part C: Progressive Deepening (10 points)
Assume you have the same setup as in Part B, except this time, you have an 5-second time-limit for
each move. Each static evaluation takes 1 second. Assume there is no time associated with building
the alpha-beta tree and no time associated with performing the alpha beta search. You are to use
progressive deepening and alpha-beta together.
Before performing alpha-beta at any level, use all you know from previous calculations to reorder the
nodes at ALL higher levels as much as possible. Naturally, for any reordering, a value...
View Full Document
This document was uploaded on 03/17/2014 for the course EECS 6.034 at MIT.
- Fall '10
- Artificial Intelligence