Efficiency of parallel exponential evolutionary programming

In this paper, we present a parallel exponential evolutionary programming (EEP) which uses a mutation based on double exponential probability distribution with positive parameter value. The distribution of random number which is used in mutation of evolutionary programming (EP) should be large in early stage and small in mid to final of evolution applying on optimization problems. In order to realize such distribution of random number, we propose EEP with parameter value of the distribution as the increasing function of evolving generation so as to decrease the variance of its distribution. In order to obtain the appropriate parameter value for a given problem, we have to test for various candidate parameter values within some range. For the purpose of this work, we adopt a parallel approach. Fortunately, EEP procedure can be easily parallelized by using only different parameter value, in order to obtain the more qualified solution applying to the given problem. Moreover, we propose a new EEP procedure with nonstrategic parameter as an extension of the parallel EEP. The experimental results show that this parallel EEP outperforms the existing EP procedures.