Beruflich Dokumente
Kultur Dokumente
(2019-20)
-------------------------------------- ----------------------------------------
DECLARATION
We the following students do hereby certify that the project report entitled, "GAME
THEORY AND ITS APPLICATION" being submitted to Atal Bihari College, Basudebpur,
Bhadrak, Odisha, for the award of Bachelor of Science, is an original peace of work has not
been submitted elsewhere for any other academic degree or diploma to this college or any
other college/university.
3. Soumyaranjan Nayak------------------------------------
4. Kalakar Nayak--------------------------------------------
5. Jyotisankar Dash-----------------------------------------
7. Sanjay Malik----------------------------------------------
8. Satyajit Barik---------------------------------------------
ACKNOWLEDGEMENT
With Limitless humility we would like to thank God the supreme and merciful for his blessing
towards us with all favourable circumstances to go through this arduous task.
We express Our Sincere and hearty gratitude to madam Sujata Mohapatra (HOD)Sr. Lecturer
Department of Math. A.B. College, Basudevpur for her constant and cordial support and
guidance during the preparation of our project. She is not only our guide and mentor but also
torch bearer in this journey.
We are also highly indebted to our lecture Mr.Parameswar Panda. (Dept. of Math.) for his
unconditional support and encouragement during our project work.
lt is our pleasure to express our sincere thanks to our parents who not only extended there
support but also extended financial help.
Finally I thanks to our groupmate for our collective co-operation and to complete the project
Work successfully.
1.
2.
3.
4.
5.
6.
7.
8.
ABSTRACT
In this article we present an overview on two person zero-sum games, which play a central
role in the development of the theory of games two person zero-sum game is a special class
of game theory on which one player’s wins that the other players loses with only two
players , it is difficult to solve two person matrix game with the order m*n (m>=3,n>=3).
The aim of the article is to determine the method on how too solve a two person matrix
game by linear programming function.
Lin-progm( ) in MATLAB with linear programming techniques on the mat lab software,
we present effective method for solving large zero-sum games problem.
The concept of game theory was developed by professor John Von Newman and Oscar
Morgenstern when they published a book “The theory of games and economic behaviour”.
Game Theory is now widely used in economic, business, administration and in many
humanity disciplines.
It is a useful scientific approach to rational decision making. When decisions have to be
made under conditions of uncertainty, two or more opponents are involved and each one
wants to optimize decision in his favours at the cost of other, game theory plays a vital role.
In game theory the two opponents are known as players.Each has finite or infinite number
of alternatives or strategies. Each player pays something to other which is called pay-off.
CHAPTER – 1(INTRODUCTION)
In his chapter it given a synoptic idea about the uses of Game Theory in various fields
of our daily life.
CHAPTER – 2
In this chapter, history, uses of Game Theory in all field being discussed. In literature
survey the books, journals of the mathematician have been reviewed The focus of different
works have been highlighted.
CHAPTER – 3
In this chapter a zero sum game and its mathematical representation of a situation in
which each participant gain or loss of utility is exactly balanced by loss or gain of utility of
other participants have been highlighted.
CHAPTER – 4
In this chapter the two person zero sum game has been discussed in which using two
players how the sum of payoffs to each player is constant for all possible outcomes for the
game are highlighted such games are sometime called constant-sum of games.
CHAPTER - 5
In this chapter finding of the work is presented with a great hope and expectation.
CONTENTS
DECLARATION
CERTIFICATE
ACKNOWLEDGEMENTS
ABSTRACT
LIST OF TABLE
LIST OF FIGURE
CHAPTER-1
INTRODUCTION
CHAPTER-2
HISTORY AND LITARTURE SURVEY
CHAPTER-3
ZERO SUM GAME
CHAPTER-4
TWO PERSON ZERO SUM GAME
CHAPTER-5
CONCLUSION/FUTURE WORK
BIBLOGRAPHY
LIST OF TABLES
JOHN VON NEUMANN
Born-28 December 1903
Died-8 February 1957
John von Neumann has played an important role in post-war economic theory. He published
“Theory of Games and Economic Behaviour” in 1944 which detailed a ground-breaking
mathematical theory of economic and social organization, based on a theory of games
strategy.
JOHN NASH
Born- 13 June 1928
Died- 23 May 2015
John Nash was American mathematician who has awarded the 1994 Nobel Prize for
economics for his landmark work, first begun in the 1950s, on the mathematics of game
theory.
LIST OF FIGURE
The idea of Pareto optimal payoff in a zero-sum game gives rise to a generalized relative
selfish rationality standard, the punishing-the-opponent standard, where both players
always seek to minimize the opponent's payoff at a favourable cost to himself rather to
prefer more than less. The punishing-the-opponent standard can be used in both zero-sum
games (i.e. warfare game, Chess) and non-zero-sum games (i.e. pooling selection games).
For two-player finite zero-sum games, the different game theoretic solution concepts of
Nash equilibrium, minimax and maximin all give the same solution, If the players are
allowed to play a mixed strategy. the game always has an equilibrium.
Applications:
Zero-sum game are found in game theory, but are less common then non-zero sum games.
Poker and gambling are popular examples of zero-sum games since the sum of the amounts
won by some players equals the combined losses of the others.
CHAPTER - 4
The first type of game that we will be looking at are those that involve only two
players. This can be a game or situation that is between two humans, a human and
a computer or human versus nature. Whoever is involved in the game will not take
away from the main goal of the game, which is to play such that they are coming out
with the best possible outcome no matter the circumstances. This is why it is
important when playing a game, to play strategically. For the most part, players will
be facing opponents who are rational in game play and hence they need to evaluate
the actions they are able to take in a game. To create an optimal outcome, players
need to look at their strategies. There are two types of strategies that we will be
studying, these include pure and mixed strategies. It is important to understand the
difference between these two types. In this chapter, the primary sources used were
[1], [3] and [4].
First we will be looking at pure strategies, which are the most basic of the two
types and then we will continue on to mixed strategies.
DEFINITION 3.1. A Pure Strategy defines the specific action a player takes,
no matter the situation of the game. Pure strategies are not random and the player
does not change their actions during the game.
Consider the situation where you are playing the simple game of rock, paper,
scissors. The rules are simple, each player has three choices to choose from in an
attempt to defeat their opponent; rock, paper or scissors. The outcomes of the game
are; paper beats rock, rock beats scissors and scissors beat paper. A player either wins,
losses or ties. If a player were to play a pure strategy of rock every time it would most
likely not be beneficial for them in this type of game. As we have stated before, it is
assumed that in a game we are playing against a rational player. If you are playing
rock every single turn, most likely your rational opponent will catch on to this pure
strategy and play accordingly, hence choosing paper. In situations or games such as
rock, paper, scissors it is beneficial for a player to play a mixed strategy, which we will
discuss in section 2 of this chapter.
DEFINITION 3.4. An outcome of a matrix game (with payoffs to the row player)
is called a saddle point if the entry at that outcome is both less than or equal to
any entry in its row, and greater than or equal to any entry in its column.
This strategy has players looking at what their best response would be to what their
opponent may play and also their own choices. If we have a payoff matrix game repre -
senting a normal form game, and we have found an element where aij is the
maximum of all of the minimums of all the rows and it is also the minimum of all of
the maximums of the columns, then is a saddle point for the game.
Chapter 4. Two Person Zero Sum Games 7
It is an easy process to find a saddle point within a game, if it does exist. A
player will begin by writing down the minimum entry for each row. From these
minimum entries choose the maximum entry, as done in the table Finding Saddle Points
in a Game. This is called the maximin. The next step is to look at the columns and write
down the maximum entry for each column. Next choose the minimum entry of these
column maximums. This is called the minimax. If the maximin of the rows and the
minimax of the columns are equal, then they appear at the saddle point strategies for
the players. If the maximin and minimax are not equal, then there is not a saddle point
in the game. In the table Finding Saddle Points in a game the saddle point occurs are
the outcome AA.
TABLE 1. Finding Saddle Points in a Game
Player 2
A B C Min
2 4 3 2
lay
1 -10 5 -10
-1 6 -8 -8
Max 2 6 5
If a saddle point exists within a payoff matrix game, then this saddle point
strategy will find what is known as the value of the game.
If the value of the game exists, then a player can guarantee themselves an outcome
of at least v. If a player does not play the saddle point strategy they are risking an
outcome that is inferior to v and their opponent can possibly have an outcome with a
value better than v. In a non-zero sum game a player would obviously want their
opponents outcome to be at max v because if it is any higher than they are losing from
their payoff of the game.
2. Mixed Strategy
Until this point we have looked solely at pure strategies, but what if there is not
a saddle point strategy or dominant strategy to a game? Then a player will want to
create what is called a mixed strategy.
DEFINITION 3.8. A plan, which involves playing a mixture of strategies according
to certain fixed probabilities is called a mixed strategy.
Instead of playing the same choice every turn, a player may want to play a mixture
of choices that will optimize their overall outcome and overall payoff while playing a
game. Now, let us suppose that we are looking at a basic 2 by 2 game, where there are
no pure strategies within the game. Hence, we want to find a mixed strategy for the
game.
In games where a player is playing a mixed strategy, they want to play a choice
x percentage of the time, which they have calculated will have the best possible outcome
of their game and another choice x — 1 percentage of the time (For a basic 2 by 2
matrix). How would we find this players mixed strategy, such that it is optimal in a
payoff matrix game? One way is to calculate the following algorithms;
Example-1
The pay-off matrix of a game is given below. Find the solution of the game to A and B.
B
I II III IV V
I -2 0 0 5 3
II 3 2 1 2 2
A
III -4 -3 0 -2 6
IV 5 3 -4 2 -6
Sol. To find the saddle point, circling the row minimum and putting squares around the
column maximums, we get the following table:
B
I II III IV V
-2 0 0 3 -2
I 5
II 3 2 1 2 2 1 v (MAX VALUE)
A
III -4 -3 0 -2 6 -4
IV
5 3 -4 2 -6 -6
COULMN
MAXIMUMS
5 3 1 5 6
v (MIN. VALUE)
Obviously, the matrix has a saddle point at the position (2,3). Thus the solution to the game
is given by
(i) The best strategy for plays A is II.
(ii) The best strategy for plays B is III.
and (iii) The value of the game is 1 for A and -1 for B.
Example-2
Solve the following game :
B
I II III
I 6 8 6
A
II 4 12 2
Sol. To find the saddle point, circling the row minimums and putting square around the
column maximums, we get the following tables:
B
I II III
I 6 8 6 6 6 Vi(MAX VALUE)
A
II 4 12 2
COULMN
6 12 6
MAXIMUMS
v (MIN. VALUE)
Obviously, the matrix have two saddle points at (1,1) and (1,3). Thus the solution to the game
is given by
(i) The best strategy for player A is I.
(ii) The best strategy for player B is II.
(iii) The value of the game is 6 for A and -6 for B.
Example-3
Solve the problem with the pay-off matrix.
Player B
1 -2 5 -3 6 7
2 4 6 8 -1 6
Player A
3 8 2 3 5 4
4 15 14 18 12 20
Sol:
Find min of each row
Row Min Value
1 -3
2 -1
3 2
4 12
Maximum of {-3, -1, 2, 12} = 12
Find max of each column
Column Max Value
I 15
II 14
III 18
IV 12
V 20
Minimum of (15, 14, 18, 12, 20) = 12
Game has saddle point. So the common value V = 12
Best strategy for players A is strategy 4.
Best strategy for players B is strategy IV.
The game is favourable for player A.
CHAPTER-5
CONCLUSION
In this paper, we examined two types of games that are involved within game theory,
zero sum and two-person zero sum games. In both of these games we studies zero-sum
and two person sum games and some of the strategies that we may apply to games to have
the best outcome in our play. We studied rational play between two players and
concluded that players should play their pure or mixed strategy that will give them
the value of the game. We then applied some of these strategies, such as the dominance
and saddle point strategies, into basic games and were able to find the value of the
games through the use of algorithms and basic steps. Next, we researched games where
we were not playing games against a rational opponent, but rather games against
nature.
Following this, we studied equilibrium and more importantly Nash Equilibrium, from
which we can take away the ability to apply mathematics with the study of human in -
teractions. Additionally we researched some important terms such as pare to optimal and
non-pare to-optimal and in order for the equilibrium to be a solution to the game it must
have been pare to-optimal.
Until the point of equilibrium, we had focused solely on two-person games. We then
expanded our research into n-person games and characteristic function form. Within this
part of our studying we looked at imputations, domination and stable sets. Lastly
in n-person games we studied the solution concept of the core. This tol d us that if we
can find an imputation within our core that does not have any imputation or coalition
that dominates it then this is the solution to our game.
Bibliography
1913-E. Zermelo provided the first theorem of game theory
asserts that chess is strictly determined
1928- John von Neumann proved the minimax theorem
1944 - John von Neumann/Oskar Morgenstem's wrote "Theory of Games and
Economic Behavior"
1950-1953, John Nash describes Nash equilibrium
1972- John Maynard Smith wrote "Game Theory and The Evolution of Fighting"