Scribd
Upload
297 views6 pages

Saddle Points

The document discusses different types of games including deterministic games with pure strategies and probabilistic games with mixed strategies. It describes payoff matrices that show the payoffs or outcomes for all combinations of player strategies. The value of a game is the result. Deterministic games may have a saddle point, where the maximin-minimax principle finds the optimal strategies for players to maximize their minimum payoff and minimize their maximum loss. Finding the saddle point involves identifying the row minimums, column maximums, and the cell where they intersect, which indicates the value and solution strategies. An example payoff matrix is provided to demonstrate this process.

Uploaded by

Yasoda Kumari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
297 views6 pages

Saddle Points

The document discusses different types of games including deterministic games with pure strategies and probabilistic games with mixed strategies. It describes payoff matrices that show the payoffs or outcomes for all combinations of player strategies. The value of a game is the result. Deterministic games may have a saddle point, where the maximin-minimax principle finds the optimal strategies for players to maximize their minimum payoff and minimize their maximum loss. Finding the saddle point involves identifying the row minimums, column maximums, and the cell where they intersect, which indicates the value and solution strategies. An example payoff matrix is provided to demonstrate this process.

Uploaded by

Yasoda Kumari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

 Determistic Games: Games with Pure

Strategies
i.e., Games in which the two players
attempt the game always with specific
strategies
 Probabilistic Games: Games with Mixed
Strategies
i.e., Games in which the two players
attempt the game probably with
different strategies
 Payoff: An outcome of a game
 Payoff Matrix: A matrix formed by the payoffs
of all combinations of strategies of the two
players.
Strategy of B1 B2 B3
Ex: A/B
A1 4 6 -2
A2 3 0 3
A3 -2 6 9
The payoffs in the above matrix are the gains
of the player A (losses of B)
Gain of A for (A1,B1) = 4 = loss of B
** Payoff Matrix is simplest toll to solve the
games.
 Value of the game: It is the result of the game
 Games with Saddle Point and solution:
If the games are determistic, then
there exists a saddle point for the
players.
The saddle point describes the
solution of the game.
The saddle point may not be unique.
The saddle point can be obtained by a
principle which is called
Maximin-Minimax principle.
 Maximin-Minimax Principle:
If the payoffs in payoff matrix are the
gains of Player A, then
Maximin principle says that the player
A always tries to maximize his
minimum gains corresponding to the
opponent strategies.
Minimax principle says that the player
B always tries to minimize his
maximum losses corresponding to the
opponent strategies.
Finding Saddle Point:
 Find row minimums -> find maximum of these
minimums. It is maximin value
 Find column maximums -> find minimum of
these maximums. It is minimax value
 Identify the pair of strategies for which
maximin=minimax.
 The point corresponding to the pair is the
saddle point.
 The pair of strategies are the pure strategies
 Value = maximin/minimax value
Note: If maximin ‡ minimax, the game an not be
determined strictly by this way. Where we will
discuss games without saddle points later
 Example:

Strategy of B1 B2 B3
A/B
A1 4 6 2
A2 5 0 -3
Solution:
A3 -2 6 1

Strategy of B1 B2 B3 Row Min


A/B
A1 4 6 2 2 -> Max
(maxmin)
A2 5 0 -3 -3
A3 -2 6 1 1
Column 5 6 2 -> Min
max (minimax)
So, maximin = minimax = 2= value
and optimum pure strategies are A2 and B3.

You might also like