Evolutionary Contribution and Problem Heuristic Information Ensemble-Based Resource Allocation for Cooperative Coevolution

Abstract

This article proposes an evolutionary contribution and problem heuristic information ensemble-based computing resource allocation (CRA) scheme for cooperative co-evolutionary algorithms (CCEAs). For problem heuristic information (HI), this article assembles the correlation sensitivity of variables in each subproblem and the dimension ratio of this subproblem; for evolutionary contribution (EC), this article assembles the historical and the current ECs of each subproblem. By assembling these two crucial factors, the devised method computes the selection probability of each subproblem and then randomly picks one subproblem by the roulette wheel selection strategy to undergo optimization in each iteration. In this way, computing resources are preferentially allocated to those subproblems with high complexity manifested by the problem HI and high fitness improvement reflected by the EC. With this method, CCEAs expectedly fully utilize the computing resources to achieve satisfactory performance in addressing large-scale optimization problems (LSOPs). By combining the devised method with six latest decomposition methods along with two evolutionary optimizers, this article has conducted experiments to compare it with seven state-of-the-art CRA methods on two popular suites of LSOPs. Experimental results have proved that the devised method outperforms the seven compared methods in helping CCEAs achieve better performance.