Many-Objective Evolutionary Algorithms Based on Coordinated Selection Strategy

He, Zhenan; Yen, Gary G.

doi:10.1109/tevc.2016.2598687

Cited by 123 publications

(41 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As introduced in Section II, RVEA [40], LWS [44], MOEA/D-SAS [45], VaEA [41], MaOEA-CSS [47], and MOEA/VAN [50] utilize the information of vector angle, while SPEA2+SDE [17], CMODE+SDE [42], and SRA [39] employ the shift-based density estimation. To the best of our knowledge, AnD is the first attempt to combine both the vector angle and the shift-based density estimation together, by taking advantage of their complementary features.…”

Section: E Discussionmentioning

confidence: 99%

See 1 more Smart Citation

AnD: A many-objective evolutionary algorithm with angle-based selection and shift-based density estimation

Liu

Wang

Huang

2020

Information Sciences

118

View full text Add to dashboard Cite

Evolutionary many-objective optimization has been gaining increasing attention from the evolutionary computation research community. Much effort has been devoted to addressing this issue by improving the scalability of multiobjective evolutionary algorithms, such as Pareto-based, decomposition-based, and indicator-based approaches. Different from current work, we propose a novel algorithm in this paper called AnD, which consists of an angle-based selection strategy and a shift-based density estimation strategy. These two strategies are employed in the environmental selection to delete the poor individuals one by one. Specifically, the former is devised to find a pair of individuals with the minimum vector angle, which means that these two individuals share the most similar search direction. The latter, which takes both the diversity and convergence into account, is adopted to compare these two individuals and to delete the worse one. AnD has a simple structure, few parameters, and no complicated operators. The performance of AnD is compared with that of seven state-of-the-art many-objective evolutionary algorithms on a variety of benchmark test problems with up to 15 objectives. The experimental results suggest that AnD can achieve highly competitive performance. In addition, we also verify that AnD can be readily extended to solve constrained many-objective optimization problems.

show abstract

Section: E Discussionmentioning

confidence: 99%

“…Other kinds of attempts have also been made to solve MaOPs with the usage of vector angle. For example, He and Yen [47] suggested an MaOEA based on coordinate selection strategy (MaOEA-CSS), in which a new diversity measure based on vector angle is designed in the mating and environmental selection.…”

Section: Preliminary Knowledge a Vector Anglementioning

confidence: 99%

AnD: A many-objective evolutionary algorithm with angle-based selection and shift-based density estimation

Liu

Wang

Huang

2020

Information Sciences

118

View full text Add to dashboard Cite

show abstract

“…On the large scale problems with hundreds of decision variables and huge amount of FEs provided, the good performance of CLIA demonstrates that the evolution pressure provided by the selection operator of cascade clustering is satisfactory, even without additional efforts on the offspring generation that may greatly boost the qualities of the population [55].…”

Section: Comprehensive Capabilities Analysesmentioning

confidence: 99%

A Many-Objective Evolutionary Algorithm With Two Interacting Processes: Cascade Clustering and Reference Point Incremental Learning

Zhao

Sun

et al. 2019

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Researches have shown difficulties in obtaining proximity while maintaining diversity for many-objective optimization problems. Complexities of the true Pareto front pose challenges for the reference vector-based algorithms for their insufficient adaptability to the diverse characteristics with no priors. This paper proposes a many-objective optimization algorithm with two interacting processes: cascade clustering and reference point incremental learning (CLIA). In the population selection process based on cascade clustering (CC), using the reference vectors provided by the process based on incremental learning, the nondominated and the dominated individuals are clustered and sorted with different manners in a cascade style and are selected by round-robin for better proximity and diversity. In the reference vector adaptation process based on reference point incremental learning, using the feedbacks from the process based on CC, proper distribution of reference points is gradually obtained by incremental learning. Experimental studies on several benchmark problems show that CLIA is competitive compared with the state-of-the-art algorithms and has impressive efficiency and versatility using only the interactions between the two processes without incurring extra evaluations.Index Terms-Clustering, incremental machine learning, interacting processes, many-objective optimization, reference vector. * Equal contribution. This is an UPDATED preprint of https: //ieeexplore.ieee.org/document/

show abstract

“…Originally, it was considered in the decision space, where by increasing the number of variables, the search space became larger, more complicated and more vulnerable to the effect of numerical rounding errors [56,64]. With the development of multi-objective optimization, scalability is also considered in the objective space with more than two objectives [30,34,37,63], which we call many-objective optimization.…”

Section: Research Issues and Problem Difficultiesmentioning

confidence: 99%

Multi-Objective Evolutionary Algorithms: Foundation, Development and Open Issues

Lam

2018

JCC

View full text Add to dashboard Cite

Abstract. Evolutionary computation (EC) has been a fascinating branch of computation inspired by a natural phenomenal of evolution. EC enables computer scientists to design effective algorithms dealing with difficult problems. This paper focuses on a special class problem called multi-objective optimization problems and evolutionary algorithms designed for it. We will overview the development of multi-objective evolutionary algorithms (MOEAs) over the years and problem difficulties and then indicate the open problems in this area. Our chief goal is to provide readers with reference material in the area of multi-objective evolutionary algorithms.

show abstract

Many-Objective Evolutionary Algorithms Based on Coordinated Selection Strategy

Cited by 123 publications

References 33 publications

AnD: A many-objective evolutionary algorithm with angle-based selection and shift-based density estimation

AnD: A many-objective evolutionary algorithm with angle-based selection and shift-based density estimation

A Many-Objective Evolutionary Algorithm With Two Interacting Processes: Cascade Clustering and Reference Point Incremental Learning

Multi-Objective Evolutionary Algorithms: Foundation, Development and Open Issues

Contact Info

Product

Resources

About