GAME PLAYING

UNIT III
Kevin Night and Elaine Rich, Nair B.
Chapter 12 : Pg No 231 - 246
 Topics
• Overview
• Minimax Search Procedure
• Adding Alpha-beta Cutoffs

CS18601 Unit 3 Game Playing - R

03/29/22 2
Jayabhaduri
 Games
• A game consists of a set of two or more players, a set of
moves for the players, and a specification of payoffs
(outcomes) for each combination of strategies.

Randomness can usually be modeled by ”hidden information” - are

called games with imperfect information

CS18601 Unit 3 Game Playing - R

03/29/22 3
Jayabhaduri
 Game as Search Problem
• Initial State: board position
and player to move
• Successor Function: returns a
list of legal (move, state) pairs
• Terminal Test: determines
when the game is over
• Utility function: Gives a
numeric value for the
terminal state

CS18601 Unit 3 Game Playing - R

03/29/22 4
Jayabhaduri
 Game Trees
• Game trees are used to
represent two-player
games.
• Alternate moves in the
game are represented by
alternate levels in the
tree (plies).
• Nodes in the tree
represent positions.
• Edges between nodes
represent moves.
• Leaf nodes represent
won, lost or drawn
positions.

CS18601 Unit 3 Game Playing

03/29/22 5
- R Jayabhaduri
 Game Trees - Structure
A strategy defines a complete plan of action for a given player.

This is an example of a
partial game tree for the
game tic-tac-toe.
Even for this simple game,
the game tree is very large.

CS18601 Unit 3 Game Playing

03/29/22 6
- R Jayabhaduri
 Assumptions
• In talking about game playing
systems, we make a number of
assumptions:
– The opponent is rational –
will play to win.
– The game is zero-sum – if
one player wins, the other
loses.
– Usually, the two players
have complete knowledge
of the game.
For games such as poker, this
is clearly not true.

CS18601 Unit 3 Game Playing

03/29/22 7
- R Jayabhaduri
 Minimax
• Minimax is a method used to
evaluate game trees.
• A static evaluator is applied to
leaf nodes, and values are
passed back up the tree to
determine the best score the
computer can obtain against a
rational opponent.
• Consider two players, MAX and
MIN. Player MAX is trying to
maximize the score and player
MIN is trying to minimize the
opponent’s score. We assume
that the players are rational.

CS18601 Unit 3 Game Playing

03/29/22 8
- R Jayabhaduri
 Minimax algorithm
• The minimax algorithm allows each player to compute their
optimal move on a game tree of alternating MAX and MIN
nodes.

CS18601 Unit 3 Game Playing - R

03/29/22 9
Jayabhaduri
 Minimax Example

CS18601 Unit 3 Game Playing - R

03/29/22 10
Jayabhaduri
 Minimax – Animated Example
Max 3 6 The computer can
obtain 6 by
choosing the right
Min 6 hand edge from the
5 3 first node.

Max 1 3 6 0 7
5

5 2 1 3 6 2 0 7

CS18601 Unit 3 Game Playing - R

03/29/22 11
Jayabhaduri
 Minimax Function

• MINIMAX-VALUE(n) =
– UTILITY(n) if n is a terminal state
– maxs  Successors(n) MINIMAX-VALUE(s) if n is a
MAX node
– mins  Successors(n) MINIMAX-VALUE(s) if n is a
MIN node

CS18601 Unit 3 Game Playing - R

03/29/22 12
Jayabhaduri
 Searching Game Trees
• Exhaustively searching a game tree is not usually a
good idea.
• Even for a game as simple as tic-tac-toe there are
over 350,000 nodes in the complete game tree.
• An additional problem is that the computer only gets
to choose every other path through the tree – the
opponent chooses the others.

CS18601 Unit 3 Game Playing - R

03/29/22 13
Jayabhaduri
 Alpha-beta Pruning
• Minimax suffers from the problem that the number of game
states it has to examine is exponential in the number of moves.
• Alpha-beta pruning - Based on the idea that if a move is clearly
bad, there is no need to follow the consequences of it
– A method for reducing the number of nodes that need to be evaluated by
only considering nodes that may be reached in game play.
– A method that can often cut off a half the game tree
• Alpha-beta pruning places bounds on the values appearing
anywhere along a path:
– 𝛼: the best (highest) value found so far for MAX
– 𝛽: the best (lowest) value found so far for MIN
– 𝛼 and 𝛽 propagate down the game tree. The value v propagates up the
game tree.

CS18601 Unit 3 Game Playing - R

03/29/22 14
Jayabhaduri
 Alpha-beta Pruning (Contd..)

CS18601 Unit 3 Game Playing - R

03/29/22 15
Jayabhaduri
 Alpha-beta Pruning Example

CS18601 Unit 3 Game Playing - R

03/29/22 16
Jayabhaduri
 α-β pruning example

CS18601 Unit 3 Game Playing - R

03/29/22 17
Jayabhaduri
 α-β pruning example
=3

alpha cutoff

CS18601 Unit 3 Game Playing - R

03/29/22 18
Jayabhaduri
 α-β pruning example

CS18601 Unit 3 Game Playing - R

03/29/22 19
Jayabhaduri
 α-β pruning example

CS18601 Unit 3 Game Playing - R

03/29/22 20
Jayabhaduri
 α-β pruning example

CS18601 Unit 3 Game Playing - R

03/29/22 21
Jayabhaduri
 Alpha Cutoff
>3 =3

8 10

What happens here? Is there an alpha cutoff?

CS18601 Unit 3 Game Playing - R

03/29/22 22
Jayabhaduri
 Beta Cutoff

=4 <4

4 >8

8  cutoff

CS18601 Unit 3 Game Playing - R

03/29/22 23
Jayabhaduri
 Example : 2 Alpha-Beta Pruning
http://people.cs.pitt.edu/~litman/courses/cs2
710/lectures/pruningReview.pdf

CS18601 Unit 3 Game Playing - R

03/29/22 24
Jayabhaduri
 CS18601 Unit 3 Game Playing - R
03/29/22 25
Jayabhaduri
 CS18601 Unit 3 Game Playing - R
03/29/22 26
Jayabhaduri
 CS18601 Unit 3 Game Playing - R
03/29/22 27
Jayabhaduri
 The α-β algorithm

cutoff

CS18601 Unit 3 Game Playing - R

03/29/22 28
Jayabhaduri
 Properties of α-β
• Effects of pruning
– Reduces the search space
– Does not affect final result

CS18601 Unit 3 Game Playing - R

03/29/22 29
Jayabhaduri
 The Idea of α-β

CS18601 Unit 3 Game Playing - R

03/29/22 30
Jayabhaduri
 The α-β Algorithm

CS18601 Unit 3 Game Playing - R

03/29/22 31
Jayabhaduri
 Static Evaluation Functions
• A static evaluator assigns a score to a position:
– High positive = computer is winning
– Zero = even game
– High negative = opponent is winning
• It is most important that a static evaluator will
give a better score to a better position – the actual
values are not so important.

CS18601 Unit 3 Game Playing - R

03/29/22 32
Jayabhaduri