Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization
Bayesian evolutionary optimization algorithms have been widely employed to solve expensive many-objective optimization problems. However, the existing approaches are generally designed for low-dimensional problems. In high-dimensional problems, the accuracy of the prediction decreases. And the acqui...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/171437 |
| _version_ | 1826112472245862400 |
|---|---|
| author | Bian, Hongli Tian, Jie Yu, Jialiang Yu, Han |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Bian, Hongli Tian, Jie Yu, Jialiang Yu, Han |
| author_sort | Bian, Hongli |
| collection | NTU |
| description | Bayesian evolutionary optimization algorithms have been widely employed to solve expensive many-objective optimization problems. However, the existing approaches are generally designed for low-dimensional problems. In high-dimensional problems, the accuracy of the prediction decreases. And the acquisition function becomes ineffective. The combination of these challenges renders existing approaches unsuitable for selecting potential individual solutions for high-dimensional many-objective optimization problems. To address these limitations, we propose a novel Entropy Search-based Bayesian Co-Evolutionary Optimization approach (ESB-CEO). With the co-evolutionary algorithm as the basic optimizer, it executes an adaptive acquisition function combining the Lp-norm and information entropy to efficiently solve computationally expensive many-objective optimization problems. Individual solutions that have a significant effect on different search stages can be effectively identified, which improves the convergence and diversity of the algorithm. Extensive experimental results based on a set of expensive multi/many-objective test problems demonstrate that the proposed approach significantly outperforms five state-of-the-art surrogate-assisted evolutionary algorithms. |
| first_indexed | 2024-10-01T03:07:34Z |
| format | Journal Article |
| id | ntu-10356/171437 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:07:34Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1714372023-10-24T08:16:28Z Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization Bian, Hongli Tian, Jie Yu, Jialiang Yu, Han School of Computer Science and Engineering Engineering::Computer science and engineering Adaptive Acquisition Function Entropy Search Bayesian evolutionary optimization algorithms have been widely employed to solve expensive many-objective optimization problems. However, the existing approaches are generally designed for low-dimensional problems. In high-dimensional problems, the accuracy of the prediction decreases. And the acquisition function becomes ineffective. The combination of these challenges renders existing approaches unsuitable for selecting potential individual solutions for high-dimensional many-objective optimization problems. To address these limitations, we propose a novel Entropy Search-based Bayesian Co-Evolutionary Optimization approach (ESB-CEO). With the co-evolutionary algorithm as the basic optimizer, it executes an adaptive acquisition function combining the Lp-norm and information entropy to efficiently solve computationally expensive many-objective optimization problems. Individual solutions that have a significant effect on different search stages can be effectively identified, which improves the convergence and diversity of the algorithm. Extensive experimental results based on a set of expensive multi/many-objective test problems demonstrate that the proposed approach significantly outperforms five state-of-the-art surrogate-assisted evolutionary algorithms. Agency for Science, Technology and Research (A*STAR) Nanyang Technological University National Research Foundation (NRF) This research is supported, in part, by National Science Foundation of China under Grants 62006143; National Science Foundation of Shandong Province (ZR2020MF152); National Science Foundation of China under Grants 61773192, 61803192 and 61773246; the National Research Foundation Singapore and DSO National Laboratories under the AI Singapore Programme (AISG Award No: AISG2-RP-2020-019); the Nanyang Assistant Professorship (NAP); and the RIE 2020 Advanced Manufacturing and Engineering (AME) Programmatic Fund (No. A20G8b0102), Singapore. 2023-10-24T08:16:27Z 2023-10-24T08:16:27Z 2023 Journal Article Bian, H., Tian, J., Yu, J. & Yu, H. (2023). Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization. Knowledge-Based Systems, 274, 110630-. https://dx.doi.org/10.1016/j.knosys.2023.110630 0950-7051 https://hdl.handle.net/10356/171437 10.1016/j.knosys.2023.110630 2-s2.0-85163454460 274 110630 en AISG2-RP-2020-019 A20G8b0102 Knowledge-Based Systems © 2023 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering::Computer science and engineering Adaptive Acquisition Function Entropy Search Bian, Hongli Tian, Jie Yu, Jialiang Yu, Han Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title | Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title_full | Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title_fullStr | Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title_full_unstemmed | Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title_short | Bayesian co-evolutionary optimization based entropy search for high-dimensional many-objective optimization |
| title_sort | bayesian co evolutionary optimization based entropy search for high dimensional many objective optimization |
| topic | Engineering::Computer science and engineering Adaptive Acquisition Function Entropy Search |
| url | https://hdl.handle.net/10356/171437 |
| work_keys_str_mv | AT bianhongli bayesiancoevolutionaryoptimizationbasedentropysearchforhighdimensionalmanyobjectiveoptimization AT tianjie bayesiancoevolutionaryoptimizationbasedentropysearchforhighdimensionalmanyobjectiveoptimization AT yujialiang bayesiancoevolutionaryoptimizationbasedentropysearchforhighdimensionalmanyobjectiveoptimization AT yuhan bayesiancoevolutionaryoptimizationbasedentropysearchforhighdimensionalmanyobjectiveoptimization |