5. Game Theory and Application Flashcards
The Prisoners Dilemma
- Both prisoner A’s and prisoner B’s best choice is to confess = In doing this they’re using their dominant strategies
- Equilibrium outcome then is (confess, confess) à this is known as Nash Equilibrium
- Neither prisoner given that the other prisoner has chosen to confess would choose to unilaterally change their mind.
- It would be better for both prisoners to stay quiet. à this is an example of social dilemma
Conventional Assumption
- Rationality: a player is rational and they seek to maximise their own payoffs
- Common knowledge: the payoffs and strategies available to the players are common knowledge
- Each player knows his/her own payoffs and strategies, and the other player’s payoffs and strategies
- Each player knows that the other player knows this and so on… - Complete information: there is no private information; all details of the game are common knowledge to all players.
Payoff Matrix
The table showing the payoffs of both players with each possible combination outcome.
Outcome
(strategy combination) of the game is a pair such as (U,R) where player A chooses U and player B chooses R.
Dominant Strategy
A strategy that is optimal (leading to higher payoff) no matter what one’s opponent does.
Equilibrium in Dominant Strategies
The outcome when both players have a dominant strategy.
Nash Equilibrium
A set of strategies, one for each player, such that no player has an incentive to unilaterally change their action.
Maximin strategies
A strategy that maximises the minimum gain that can be earned (dominant strategies are also maximin strategies)
Games of coordination
Simultaneous play games in which the payoffs to the players are largest when they coordinate their strategies e.g. prisoners dilemma.
- Neither know what the other is doing –> players need to move fast to enable coordination
- in practice, games played more than once: players know moves from previous periods but not within any one period because they’re all simultaneous move.
Tit-for-Tat Strategy
strategy in which players respond in kind to an opponent’s previous play, cooperating with cooperative opponents and retaliating against uncooperative ones.
Infinitely repeated games
Cooperation is a rational response to the t-f-t strategy
Finitely repeated games
Cooperation is no longer a rational response- backward induction argument (start from the end and work backwards)
Sequential Games
1 player moves first then the other chooses their strategy will full knowledge of the first players choice : represented through tree diagrams
Pure Strategies
Strategies where a player makes a specific choice or action with certainty
Games of Competition
Simultaneous play games where any increase in the payoff to one player is exactly offset by the decrease in the payoff to the other player