Solving large-scale many-objective optimization problems by covariance matrix adaptation evolution strategy with scalable small subpopulations

Chen，Huangke1; Cheng，Ran2; Wen，Jinming3; Li，Haifeng4; Weng，Jian3

2020

10.1016/j.ins.2018.10.007

INFORMATION SCIENCES   影响因子和分区

0020-0255

1872-6291

Despite the recent development in evolutionary multi- and many-objective optimization, the problems with large-scale decision variables still remain challenging. In this work, we propose a scalable small subpopulations based covariance matrix adaptation evolution strategy, namely S-CMA-ES, for solving many-objective optimization problems with large-scale decision variables. The proposed S-CMA-ES attempts to approximate the set of Pareto-optimal solutions using a series of small subpopulations instead of a whole population, where each subpopulation converges to only one solution. In the proposed S-CMA-ES, a diversity improvement strategy is designed to generate and select new solutions. The performance of S-CMA-ES is compared with five representative algorithms on 36 test instances with 5–15 objectives and 500–1500 decision variables. The empirical results demonstrate the superiority of the proposed S-CMA-ES.

CMA-ES Evolutionary algorithm Large-scale multi-objective optimization Many-objective optimization Scalable populations

SCI ; EI

英语

通讯

Shenzhen Peacock Plan[KQTD2016112514355531]

Computer Science

Computer Science, Information Systems

WOS:000494883700029

ELSEVIER SCIENCE INC

20184205959358

Decision making ; Evolutionary algorithms ; Multiobjective optimization ; Pareto principle

Management:912.2 ; Mathematics:921 ; Optimization Techniques:921.5

COMPUTER SCIENCE

2-s2.0-85054820433

Scopus

1.College of Systems EngineeringNational University of Defense Technology,Changsha,410073,China
2.Shenzhen Key Laboratory of Computational IntelligenceUniversity Key Laboratory of Evolving Intelligent Systems of Guangdong ProvinceDepartment of Computer Science and EngineeringSouthern University of Science and Technology,Shenzhen,518055,China
3.College of Information Science and TechnologyJinan University,Guangzhou,510632,China
4.School of Geosciences and Info-PhysicsCentral South University,Changsha,410004,China

Chen，Huangke,Cheng，Ran,Wen，Jinming,et al. Solving large-scale many-objective optimization problems by covariance matrix adaptation evolution strategy with scalable small subpopulations[J]. INFORMATION SCIENCES,2020,509:457-469.

Chen，Huangke,Cheng，Ran,Wen，Jinming,Li，Haifeng,&Weng，Jian.(2020).Solving large-scale many-objective optimization problems by covariance matrix adaptation evolution strategy with scalable small subpopulations.INFORMATION SCIENCES,509,457-469.

Chen，Huangke,et al."Solving large-scale many-objective optimization problems by covariance matrix adaptation evolution strategy with scalable small subpopulations".INFORMATION SCIENCES 509(2020):457-469.