Theoretical Analysis and Experimental Evidence of Opposite-Center Learning
The main motivation of this paper is to discuss some theoretical details of opposite-center learning (OCL) and further validate its effectiveness for the optimization problems. In order to reveal the strong flexibility of its definition, two analytical solutions of opposite-center point are deduced...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2018-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8263159/ |
| _version_ | 1831672367676391424 |
|---|---|
| author | Qingzheng Xu Hongpei Xu Wei Wu |
| author_facet | Qingzheng Xu Hongpei Xu Wei Wu |
| author_sort | Qingzheng Xu |
| collection | DOAJ |
| description | The main motivation of this paper is to discuss some theoretical details of opposite-center learning (OCL) and further validate its effectiveness for the optimization problems. In order to reveal the strong flexibility of its definition, two analytical solutions of opposite-center point are deduced for 1-D case. In order to reduce its computational complexity for higher dimensions, several termination criterion of iterative process are discussed thoughtfully and then a simple and efficient criterion is found when considering both the algorithm performance and computation cost. Moreover, a uniform evaluation approach to compute an evaluation function is proposed and then different opposition strategies can be compared easily by means of the mathematical expectation of these functions. To further verify its practical performance, OCL mechanism is embedded into differential evolution (DE) for population initialization and generation jumping and opposite-center DE is proposed. Simulation results demonstrate the strong exploitation ability of OCL. The obtained results also confirm a good tradeoff of solution accuracy and convergence speed in solving various function optimizations. |
| first_indexed | 2024-12-19T23:51:44Z |
| format | Article |
| id | doaj.art-e13991d298984b0b82f1149afe0f30d0 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T23:51:44Z |
| publishDate | 2018-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-e13991d298984b0b82f1149afe0f30d02022-12-21T20:01:08ZengIEEEIEEE Access2169-35362018-01-016349483496610.1109/ACCESS.2018.27949618263159Theoretical Analysis and Experimental Evidence of Opposite-Center LearningQingzheng Xu0https://orcid.org/0000-0001-8212-1073Hongpei Xu1Wei Wu2School of Information Engineering, Xijing University, Xi’an, ChinaFaculty of Science, University of Amsterdam, Amsterdam, The NetherlandsSchool of Information Engineering, Xijing University, Xi’an, ChinaThe main motivation of this paper is to discuss some theoretical details of opposite-center learning (OCL) and further validate its effectiveness for the optimization problems. In order to reveal the strong flexibility of its definition, two analytical solutions of opposite-center point are deduced for 1-D case. In order to reduce its computational complexity for higher dimensions, several termination criterion of iterative process are discussed thoughtfully and then a simple and efficient criterion is found when considering both the algorithm performance and computation cost. Moreover, a uniform evaluation approach to compute an evaluation function is proposed and then different opposition strategies can be compared easily by means of the mathematical expectation of these functions. To further verify its practical performance, OCL mechanism is embedded into differential evolution (DE) for population initialization and generation jumping and opposite-center DE is proposed. Simulation results demonstrate the strong exploitation ability of OCL. The obtained results also confirm a good tradeoff of solution accuracy and convergence speed in solving various function optimizations.https://ieeexplore.ieee.org/document/8263159/Opposite-center learningopposition-based learningdifferential evolutionsamplingcontinuous function optimization |
| spellingShingle | Qingzheng Xu Hongpei Xu Wei Wu Theoretical Analysis and Experimental Evidence of Opposite-Center Learning IEEE Access Opposite-center learning opposition-based learning differential evolution sampling continuous function optimization |
| title | Theoretical Analysis and Experimental Evidence of Opposite-Center Learning |
| title_full | Theoretical Analysis and Experimental Evidence of Opposite-Center Learning |
| title_fullStr | Theoretical Analysis and Experimental Evidence of Opposite-Center Learning |
| title_full_unstemmed | Theoretical Analysis and Experimental Evidence of Opposite-Center Learning |
| title_short | Theoretical Analysis and Experimental Evidence of Opposite-Center Learning |
| title_sort | theoretical analysis and experimental evidence of opposite center learning |
| topic | Opposite-center learning opposition-based learning differential evolution sampling continuous function optimization |
| url | https://ieeexplore.ieee.org/document/8263159/ |
| work_keys_str_mv | AT qingzhengxu theoreticalanalysisandexperimentalevidenceofoppositecenterlearning AT hongpeixu theoreticalanalysisandexperimentalevidenceofoppositecenterlearning AT weiwu theoreticalanalysisandexperimentalevidenceofoppositecenterlearning |