An improved self-adaptive differential evolutionary algorithm with population reduction

作者：Yang Ming^*; Cai Zhihua; Li Changhe

来源：International Journal of Advancements in Computing Technology, 2012, 4(15): 57-65.

DOI：10.4156/ijact.vol4.issue 15.7

• Computational resources
• Convergence speed
• DE algorithms
• Difference vectors
• Differential Evolution
• Differential evolutionary algorithm
• Population diversity
• Population sizes
• Self adaptation
• Self adaptive control
• Self-adaptive

Reducing population size could save computational resources and hence accelerate its convergence speed. This is beneficial to algorithms for optimizing problems which need expensive evaluations. But population reduction may reduce the population diversity, and algorithm will result in prematurity due to bad population diversity. In this paper, we research the effect of population reduction on self adaptive DE algorithm, and embed it into jDE algorithm. The new algorithm combines the population reduction and self-adaptive control parameters F and CR. When population needs to reduce, delete the individuals with large-step difference vectors, so it can remain the ones with small-step difference vectors as a new population. The results show that the proposed algorithm can get better results on average, and the convergence becomes faster and faster as each population reduction.

• 出版日期2012