Tag Archives: game theory

Rats can learn how to win at the Prisoner’s Dilemma

Some scientists created a Prisoner’s Dilemma for rats. And the rats did well!

The experimental rat could then decide whether to cooperate with the stooge rat, or go for the largest food payout by defecting.

The results showed that the rats quickly figured out their opponent’s strategy. For example, if the experimental rat defected, the stooge playing a tit-for-tat strategy would defect on the next trial. Rather than continually going after the high food reward, the experimental rat fell in line and quickly started cooperating again, avoiding a continuous cycle of defection. In fact, when competing against a tit-for-tat opponent, the rats cooperated about 60% of the time.

Via Boing Boing

Also tagged , | Leave a comment

Health Care Reform game theory

Jonathan Chait:

Democrats ought to learn the correct lesson, which is that reciprocity is generally the majority party's best approach. A strategy of partisan cohesion is best met with a counter-strategy of partisan cohesion. Demonstrating their willingness to hold together in the face of united opposition has softened the opposition and made bipartisanship possible. Which is to say, if Democrats had abandoned health care reform on the grounds that it lacked bipartisan support, then the prospects for bipartisan support in the future would have been dimmer than they are. If some Republicans are willing to negotiate on financial regulation, as they appear to be, Democrats ought to reciprocate.

In other words, the Democrats should employ the tit-for-tat strategy.

Via Andrew Sullivan

Also tagged , | Leave a comment

A Demonstration of the Iterated Prisoner’s Dilemma

 A Demonstration of the Iterated Prisoners Dilemma

View in the full browser

Explanation:
The Prisoner’s Dilemma is a logical exercise where two players are given the choice to cooperate with or defect against one another. The highest score a player can receive is 5 points, which he earns when he defects while his opponent cooperates. In this scenario, his opponent receives no points. If both players defect, then each receives 1 point, and if both cooperate, they receive 3 points. The Iterated Prisoner’s Dilemma (IPD) repeats this exercise many times over, giving each player a chance to react to the choices of his opponent. This sets up the dilemma: in each isolated round, it’s in the best interest of the player to defect. But if this leads both players to repeatedly defect against each other, then they will both score fewer points than if they repeatedly cooperated with each other.

In this demonstration, you can pit different strategies of play against one another in a tournament. By setting the numbers next to the names of the different agent types, you control the composition of your IPD tournament. When you hit the Play button, every agent you chose plays the same number of rounds of the Prisoner’s Dilemma against every other agent. The table displays the total results from all of these matchups.

This demonstration is based on Robert Axelrod’s original IPD tournaments as well as Chris Cook’s own demo of the IPD. On Wikipedia, you can read more about the Prisoner’s Dilemma and the Evolution of Cooperation. You can also download Chris Cook’s IPD demo. It has several features that this demo doesn’t.

Also tagged , , | 1 Comment