A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition
In the field of many-objective evolutionary optimization algorithms (MaOEAs), how to maintain the balance between convergence and diversity has been a significant research problem. With the increase of the number of objectives, the number of mutually nondominated solutions increases rapidly, and mul...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/2/413 |
| _version_ | 1797439005778771968 |
|---|---|
| author | Yizhang Xia Jianzun Huang Xijun Li Yuan Liu Jinhua Zheng Juan Zou |
| author_facet | Yizhang Xia Jianzun Huang Xijun Li Yuan Liu Jinhua Zheng Juan Zou |
| author_sort | Yizhang Xia |
| collection | DOAJ |
| description | In the field of many-objective evolutionary optimization algorithms (MaOEAs), how to maintain the balance between convergence and diversity has been a significant research problem. With the increase of the number of objectives, the number of mutually nondominated solutions increases rapidly, and multi-objective evolutionary optimization algorithms, based on Pareto-dominated relations, become invalid because of the loss of selection pressure in environmental selection. In order to solve this problem, indicator-based many-objective evolutionary algorithms have been proposed; however, they are not good enough at maintaining diversity. Decomposition-based methods have achieved promising performance in keeping diversity. In this paper, we propose a MaOEA based on indicator and decomposition (IDEA) to keep the convergence and diversity simultaneously. Moreover, decomposition-based algorithms do not work well on irregular PFs. To tackle this problem, this paper develops a reference-points adjustment method based on the learning population. Experimental studies of several well-known benchmark problems show that IDEA is very effective compared to ten state-of-the-art many-objective algorithms. |
| first_indexed | 2024-03-09T11:46:36Z |
| format | Article |
| id | doaj.art-82258073b3054bcabfb5f32ee939efd7 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T11:46:36Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-82258073b3054bcabfb5f32ee939efd72023-11-30T23:21:52ZengMDPI AGMathematics2227-73902023-01-0111241310.3390/math11020413A Many-Objective Evolutionary Algorithm Based on Indicator and DecompositionYizhang Xia0Jianzun Huang1Xijun Li2Yuan Liu3Jinhua Zheng4Juan Zou5School of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaSchool of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaSchool of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaSchool of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaSchool of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaSchool of Cyberspace Science, Xiangtan University, Xiangtan 411105, ChinaIn the field of many-objective evolutionary optimization algorithms (MaOEAs), how to maintain the balance between convergence and diversity has been a significant research problem. With the increase of the number of objectives, the number of mutually nondominated solutions increases rapidly, and multi-objective evolutionary optimization algorithms, based on Pareto-dominated relations, become invalid because of the loss of selection pressure in environmental selection. In order to solve this problem, indicator-based many-objective evolutionary algorithms have been proposed; however, they are not good enough at maintaining diversity. Decomposition-based methods have achieved promising performance in keeping diversity. In this paper, we propose a MaOEA based on indicator and decomposition (IDEA) to keep the convergence and diversity simultaneously. Moreover, decomposition-based algorithms do not work well on irregular PFs. To tackle this problem, this paper develops a reference-points adjustment method based on the learning population. Experimental studies of several well-known benchmark problems show that IDEA is very effective compared to ten state-of-the-art many-objective algorithms.https://www.mdpi.com/2227-7390/11/2/413evolutionary algorithmmany-objective optimizationreference point adjustmentlearning population |
| spellingShingle | Yizhang Xia Jianzun Huang Xijun Li Yuan Liu Jinhua Zheng Juan Zou A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition Mathematics evolutionary algorithm many-objective optimization reference point adjustment learning population |
| title | A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition |
| title_full | A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition |
| title_fullStr | A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition |
| title_full_unstemmed | A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition |
| title_short | A Many-Objective Evolutionary Algorithm Based on Indicator and Decomposition |
| title_sort | many objective evolutionary algorithm based on indicator and decomposition |
| topic | evolutionary algorithm many-objective optimization reference point adjustment learning population |
| url | https://www.mdpi.com/2227-7390/11/2/413 |
| work_keys_str_mv | AT yizhangxia amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT jianzunhuang amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT xijunli amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT yuanliu amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT jinhuazheng amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT juanzou amanyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT yizhangxia manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT jianzunhuang manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT xijunli manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT yuanliu manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT jinhuazheng manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition AT juanzou manyobjectiveevolutionaryalgorithmbasedonindicatoranddecomposition |