Description
This work aims to affirm the evolution of cooperation in a competitive population. In any complex and dynamic ecological system, agents in a population interacting for first time find it difficult to trust each other. However, when agents have an opportunity to have repeated interactions, the decision of trust can be made based on outcomes of previous interactions. The Iterated Prisoner's Dilemma is an ideal problem which is used to study the decision making of interacting agents in a population and emerging behavior of population as a result of these interactions. In nature, living organisms undergo changes over the generations to adapt to an ecosystem and eventually adopt ways to survive in a competitive environment. There are different factors which affect the evolution of living organisms, for example: genetic changes. Based on such factors a population as a whole exhibits certain dynamics, which might lead to survival or extinction of agents within the population. This work investigates- (1) The basic traits required for survival of agents in a competitive environment, (2) The factors affecting the evolution of a population, (3) Emerging behavior in evolution of a population. The multi-player The Iterated Prisoner's Dilemma with a three round history of players is used for the purposes of developing a simulation environment. Genetic algorithms are used as an evolution technique. Results of simulations carried out suggest cooperation is the dominant behavior of a population which evolved successfully as opposed to a behavior of exploitation of agents, which proved to be a stumbling block in the successful evolution of a population.
