3 ms·
The iterated prisoner's dilemma's optimal solution is tit-for-tat, which is a "nice" strategy in that it never defects first. In game-theory tournaments where
by nostrademons 11y ago
The iterated prisoner's dilemma's optimal solution is tit-for-tat, which is a "nice" strategy in that it never defects first. In game-theory tournaments where bots play off an iterated prisoner's dilemma against multiple opponents, the "naughty" strategies get crushed because they eventually encounter a provocable strategy and get caught in a defect loop, running themselves into the ground while the "nice" strategies dominated. This is the situation closest to the startup ecosystem. See eg. The Evolution of Cooperation:
https://en.wikipedia.org/wiki/The_Evolution_of_Cooperation#Axelrod.27s_tournaments https://en.wikipedia.org/wiki/The_Evolution_of_Cooperation#A...