.png){kind=icon}
返回
A strengthened diversity indicator and reference vector-based evolutionary algorithm for many-objective optimization
.png){kind=icon}
PRE
AI
DOI：10.1007/s00500-021-05981-1.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
As pointed out in recent studies, most evolutionary algorithms have shown their promise in dealing with many-objective optimization problems (MaOPs). However, the ability to balance convergence and diversity and the scalability of objectives are still far from perfect. To address these issues, this paper proposes a strengthened diversity indicator and reference vector-based evolutionary algorithm for many-objective optimization, termed SDIEA. In the proposed SDIEA, a new adaptive indicator is proposed, namely, strengthened diversity indicator (SDI). The proposed SDI is adopted to effectively emphasize the pressure of convergence and diversity in the many-objective space, by preserving the achievement scalarizing function (ASF) metric and considering angle information between individuals and reference vectors with the number of objectives and generations. Moreover, in order to further enhance the diversity management in global space, the reference vectors are used to guide population partition selection. Finally, the selected solutions based on SDI and reference vector are merged for the next generation. Experimental results show that the proposed SDIEA achieves competitive performance on MaF and WFG, compared with several state-of-the-art algorithms. Additionally, SDIEA obtains good performance on 20-objective test instances and still has the potential to strive for scalability of higher objectives.
Keyword:
Evolutionary algorithms
Many-objective optimization
Scalability
Strengthened diversity indicator
Reference vector
期刊
IF:
2.5
论文数:
1.0W
被引数:
2.1W
机构
引用论文
.png){kind=icon}
A two-stage R2 indicator based evolutionary algorithm for many-objective optimization一种基于两阶段R2指标的多目标优化进化算法
OAAI
OAAI分享收藏PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective OptimizationPlatEMO: 用于进化多目标优化的MATLAB平台OAAI
OAAI分享收藏Uncertainty-Aware Online Scheduling for Real-Time Workflows in Cloud Service Environment云服务环境下不确定性感知的实时工作流在线调度
PREAI
分享收藏A Strengthened Dominance Relation Considering Convergence and Diversity for Evolutionary Many-Objective Optimization进化多目标优化中考虑收敛性和多样性的增强优势关系OAAI
OAAI分享收藏
查看更多内容