A fast multi-objective evolutionary algorithm based on a tree structure
A fast multi-objective evolutionary algorithm based on a tree structure
复制标题
DOI：
10.1016/j.asoc.2009.08.018
复制
发表时间：
2010-03
期刊：
影响因子：
--
通讯作者：
C. Shi;Zhenyu Yan;Zhongzhi Shi;Lei Zhang
中科院分区：
文献类型：
--
作者：
C. Shi;Zhenyu Yan;Zhongzhi Shi;Lei Zhang
本文提出了一种基于树结构的多目标优化快速进化算法。该树结构称为支配树（DT），能够有效地保留个体之间必要的帕累托支配关系，隐含地包含密度信息，并显着减少个体之间的比较次数。基于支配树的进化算法（DTEA）将收敛策略和多样性策略集成到DT中，并采用基于DT的淘汰策略，在不增加时间和空间成本的情况下实现精英主义并保留种群多样性。数值实验表明，DTEA比SPEA2、NSGA-II和NSGA-II的改进版本快得多，而其解质量与SPEA2和NSGA-II具有竞争力。
This paper proposes a fast evolutionary algorithm based on a tree structure for multi-objective optimization. The tree structure, named dominating tree (DT), is able to preserve the necessary Pareto dominance relations among individuals effectively, contains the density information implicitly, and reduces the number of comparisons among individuals significantly. The evolutionary algorithm based on dominating tree (DTEA) integrates the convergence strategy and diversity strategy into the DT and employs a DT-based eliminating strategy that realizes elitism and preserves population diversity without extra time and space costs. Numerical experiments show that DTEA is much faster than SPEA2, NSGA-II and an improved version of NSGA-II, while its solution quality is competitive with those of SPEA2 and NSGA-II.