atlaua / prisoners-dilemma

Iterated Prisoners Dilemma Simulator

Prisoners Dilemma Simulator

This is a simple Prisoners' Dilemma simulator, written in Haskell and published under the MIT license.

Overview

• Pd.Master includes the main game logic. Its interface consists of play2, which simulates an iterated prisoners' dilemma with two agents, and playN, which takes n agents and simulates the same for each possible pair, summing up the scores.

• Pd.Agents has some example agents, namely Always Cooperate, Always Defect and Tit for Tat.

Usage

So far, there is no user interface, you'll have to use ghci:

λ> :l Pd/Main.hs 
[1 of 4] Compiling Pd.Types         ( Pd/Types.hs, interpreted )
[2 of 4] Compiling Pd.Master        ( Pd/Master.hs, interpreted )
[3 of 4] Compiling Pd.Agents        ( Pd/Agents.hs, interpreted )
[4 of 4] Compiling Main             ( Pd/Main.hs, interpreted )
Ok, modules loaded: Pd.Agents, Pd.Master, Main, Pd.Types.
λ> play2 1 (alwaysD, alwaysD)
(1,1)
λ> play2 1 (alwaysD, alwaysC)
(3,0)
λ> play2 10 (alwaysD, alwaysC)
(30,0)
λ> playN 1 [alwaysD, alwaysD]
[1,1]
λ> playN 1 [alwaysD, alwaysC, titForThat]
[6,2,2]
λ> playN 10 [alwaysD, alwaysC, titForThat]
[42,20,29]

Todo

• Allow alternative payoff matrices.
• Hide the plumbing, create a proper interface.