The Israel / Palestinian question reminds me strongly about one of the first examples of computational game theory that they came up with in the late 60s. They had just emerged from the Cuban Missile crisis, and the idea was that computers might be able to compete against each other in a known problem which required co-operation. They chose the Prisoner Dilemma: en.wikipedia.org/wiki/Prisoners_dilemma . In the reference example, two groups are caught by police and isolated. The police believe one of them have committed a crime: If A and B betray each other, they each spend 2 years in prison. If one betrays the other, who then remains quiet, then the betrayer goes free and the silent spends 3 years in prison. If both stay silent, then both spend 1 year in prison on a lesser charge. After years of man effort and competition the surprising result came out that the most successful way of reducing the penalties over multiple iterations was the age old eye for an eye method. By punishing the opponent should he punish you, and leaving him alone otherwise, you create a byplay that should produce a lean outcome... IF, and only IF, both groups can learn and were not from a revenge based morality. Should either party have a revenge based morality then it produces a defective outcome in eye for an eye where everyone get punished... all the time... The researchers noted this might be a problem for certain racial groups. So what does this mean? Well there is an evidence based approach saying that should you start punishing your neighbour they will respond... And this will continue if youre into revenge... Nothing gets learnt until one side stops and the other also does...
Posted on: Fri, 11 Jul 2014 02:59:07 +0000
Trending Topics
Recently Viewed Topics
© 2015