Game Thoery

Game Theory
A Game
A general situation of ….
conflict and competition
in which two or more competitors
are engaged in decision-making activities
in anticipation of certain outcomes over time.
Competitive situations
• Buyer-seller price negotiations
• Media selection to improve market share
• Union-management wage/ bonus negotiations
• Bidding for oil contract
• Pricing decisions
Terminology
• Players / competitors : The participants to the
game who act as decision makers
– Individuals / group /firms
• Strategies: A finite number of possible courses
of action available to a player
– Pure strategy
– Mixed strategy
Terminology
• Play: A play occurs when each player selects
one of their strategies.
– Assumptions:
– The choice of strategy by players is made
simultaneously
– No player knows the choice of his opponent(s)
until he has decided on his own
Terminology
• Payoff : Every combination of strategies of
players determines an outcome called payoff.
– If a gain of one player is exactly equal to a loss of
their opponent , its called Zero-sum game.
Terminology
• Payoff matrix : The gains resulting from a game
if presented in the form of a table is called
Payoff matrix.
Payoff matrix : Example
Player B
I II III
I 6 2 3
Player A
2  1  3
 
II 5 4 5 
III
Terminology
• A particular strategy (or complete plan) by
which a player optimizes his gains or losses
without knowing the competitor’s strategies
is called optimal strategy.
• The expected outcome per play when players
follow their optimal strategy is called the
value of the game.
Terminology
• Maximin:
– A maximum element among the row
minima
• Minimax :
– A minimum element among the row
maxima
Terminology
• Saddle Point :
– A saddle point is that element of the
matrix which represents the “maximin”
value of the player and “minimax” value
of their opponent.
Saddle Point
Player B
I II III
I 6 2 3
2  1  3
Player
 
 5 4 5 
A
II  
Saddle point
III
• Games
– Strategy
• Pure
• Mixed
– Players
• Two person
• Multiple person
TWO PERSON ZERO-SUM GAME
Assumptions
• Each player has a finite number of possible
courses of action. The list may not include
same number of choices for each player.
• Player A attempts to maximize gains and
player B attempts to minimize losses.
• The decisions of both players are made
individually prior to the play with no
communication between them.
Assumptions
• The decisions are made simultaneously and
also announced simultaneously so that
neither player has an advantage resulting from
direct knowledge of the other player’s
decision.
• Both the players know not only their own

possible payoffs but also of the other player.
Example
B
I II III IV
I 4 -2 -4 -1
II 3 1 -1 2
A III 2 3 -2 -2
IV -1 -3 -3 1
V -3 2 -2 -3
Solving the Game
• A is maximizing player
– Objective : To choose maximum payoff
• B is minimizing player
– Objective : To minimize A’s gain
A’s Argument
A’s Choice B’s Choice A’s gain or loss
i=I j = III -4
i = II j = III -1
i = III j = III or IV -2
i = IV j = II or III -3
i =V j = III -3
Will try to maximize the least gain i.e. Max. of min.

B’s Argument
B’s Choice A’s Choice B’s gain or loss
j=I i=I 4
j = II i = III 3
j = III i = III or IV -1
j = IV i = II or III 2
Will try to minimize the max. gain i.e. Min. of max.

Observation
• maximin aij = minmax aij = −1
• Since this is a saddle point, strategies
corresponding to saddle point are optimal
• Hence, A chooses strategy i =II and B chooses
strategy j=III
Mixed Strategy Games
• Games where there are no saddle points are
called Mixed Strategy Games
• Each strategy of a player will be selected
based on some probability p
• Types:
– 2 × 2 game
– 2 × n rectangular game
– m × 2 rectangular game
2 × 2 Game
Player B
Player A
 a11 a12 
a 
a22 
 21
2 × 2 Game
• The probabilities for optimal strategies are:
p1 = and p2 = 1-p1
q1 = and q2 = 1-q1
• Value of the game :

V=
2 × 2 Game Example
Player B
 1 3
Player A
10 2
 
Dominance Rule : Example
B
b1 b2 b3
a1 12 -8 -2
A
a2 6 7 3
a3 -10 -6 -2
Dominance Rule : Example : Step 1
B
b1 b2 b3
a1 12 -8 -2
A
a2 6 7 3
a3 -10 -6 -2
a2j > a3j . Hence a2 is dominant strategy w.r.t. a3

B
b1 b2 b3
a1 12 -8 -2
A
a2 6 7 3
bi3< bi1. Hence b3 is dominant strategy w.r.t. b1

B
b2 b3
a1 -8 -2
A
a2 7 3
a2j>a1j . Hence a2 is dominant strategy w.r.t. a1

B
b2 b3
A a2 7 3
b3 < b2 . Hence b3 is dominant strategy w.r.t. b2

Dominance Rule : Example: Step 5
B
b3
A a2 3
Solution :
A chooses a2 and B chooses b3
Value of the game = 3
Dominance Rule : Example :
Mixed Strategy
B
b1 b2 b3
a1 9 8 -7
A
a2 3 -6 4
a3 6 7 -7
RECTANGULAR GAMES
2 × n Games: Example
B
b1 b2 b3 b4
A a1 8 5 -7 9
a2 -6 6 4 -2
m × 2 Games: Example
B
b1 b2
a1 -7 6
A a2 7 -4
a3 -4 -2
a4 8 -6

Game Thoery

Uploaded by

Copyright:

Available Formats

Game Thoery

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Game Thoery

Uploaded by

Copyright:

Available Formats

Game Theory

• Both the players know not only their own

Will try to maximize the least gain i.e. Max. of min.

Will try to minimize the max. gain i.e. Min. of max.

• Value of the game :

a2j > a3j . Hence a2 is dominant strategy w.r.t. a3

bi3< bi1. Hence b3 is dominant strategy w.r.t. b1

a2j>a1j . Hence a2 is dominant strategy w.r.t. a1

b3 < b2 . Hence b3 is dominant strategy w.r.t. b2

You might also like