Ranking Model Assisted Mixed Integer Evolution Strategies

Although mixed-integer evolution strategies (MIES) have been successfully ap-plied to optimization of mixed-integer problems, they may encounter challenges when fitness evaluations are time consuming. In this paper, we propose to use two different metamodels to assisted MIES. The first one is radial-basis-function network (RBFN) trained based on the rank correlation coefficient distance metric. For the distance metric of the RBFN, we modified a heterogeneous metric (HEOM) by multiplying the weight for each dimension. Kendall rank correlation Coefficient (RCC) is adopted to measure the degree of rank correlation between the fitness and each variable. The higher the rank similarity with fitness, the greater the weight one variable will be given. Meanwhile, we tried to adopt sorting algorithm, which usually apply to learning to rank problem, to assisted MIES, and find out if the sorting algorithm was efficient for MIES. RankBoost was chosen as ranking model. Experimental results show the efficacy of the MIES assisted by the metamodels we propose.