Adaptive Parameter Selection for Strategy Adaptation in Differential Evolution for Continuous Optimization

In order to automatically select the most suitable strategy for a specific problem without any prior knowledge, in this paper, we present an adaptive parameter selection technique for strategy adaptation in differential evolution (DE). First, a simple strategy adaptation mechanism is employed to implement the adaptive strategy selection in DE. Then, the probabilitymatching-based adaptive parameter selection method is proposed to select the best parameter of the strategy adaptation mechanism; in this way, it can accelerate the strategy adaptation mechanism to choose the most suitable strategy while solving a problem. To evaluate the performance of our approach, thirteen widely used benchmark functions are chosen as the test suite. The performance of our approach is compared with other DE variants, including two recently proposed DE with strategy adaptation. The results indicate that our approach is highly competitive to the compared algorithms. In addition, compared with the classical DE algorithms with single strategy, our method obtains better results in terms of the quality of the final solutions and the convergence speed.