Does extra genetic diversity maintain escalation in a co-evolutionary arms race

摘要: In evolutionary computation (EC), genetic diversity (or its absence) gets the credit blame) for a multitude of effects — and so mutation operators, population initialization, even pseudo-random number generators, all get probed prodded to improve diversity. This paper demonstrates how extra initial can appear cause improvements in performance coevolutionary learning, but true is unforeseen problem-specific representation. The learning task considered this variation on game Iterated Prisoner’s Dilemma (IPD): here players have fine-grained range intermediate choices between full cooperation defection.