An adaptive multitasking optimization algorithm based on population distribution
Evolutionary multitasking optimization (EMTO) handles multiple tasks simultaneously by transferring and sharing valuable knowledge from other relevant tasks. How to effectively identify transferred knowledge and reduce negative knowledge transfer are two key issues in EMTO. Many existing EMTO algori...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2024-01-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2024107?viewType=HTML |
| _version_ | 1797303983993257984 |
|---|---|
| author | Xiaoyu Li Lei Wang Qiaoyong Jiang Qingzheng Xu |
| author_facet | Xiaoyu Li Lei Wang Qiaoyong Jiang Qingzheng Xu |
| author_sort | Xiaoyu Li |
| collection | DOAJ |
| description | Evolutionary multitasking optimization (EMTO) handles multiple tasks simultaneously by transferring and sharing valuable knowledge from other relevant tasks. How to effectively identify transferred knowledge and reduce negative knowledge transfer are two key issues in EMTO. Many existing EMTO algorithms treat the elite solutions in tasks as transferred knowledge between tasks. However, these algorithms may not be effective enough when the global optimums of the tasks are far apart. In this paper, we study an adaptive evolutionary multitasking optimization algorithm based on population distribution information to find valuable transferred knowledge and weaken the negative transfer between tasks. In this paper, we first divide each task population into K sub-populations based on the fitness values of the individuals, and then the maximum mean discrepancy (MMD) is utilized to calculate the distribution difference between each sub-population in the source task and the sub-population where the best solution of the target task is located. Among the sub-populations of the source task, the sub-population with the smallest MMD value is selected, and the individuals in it are used as transferred individuals. In this way, the solution chosen for the transfer may be an elite solution or some other solution. In addition, an improved randomized interaction probability is also included in the proposed algorithm to adjust the intensity of inter-task interactions. The experimental results on two multitasking test suites demonstrate that the proposed algorithm achieves high solution accuracy and fast convergence for most problems, especially for problems with low relevance. |
| first_indexed | 2024-03-08T00:02:48Z |
| format | Article |
| id | doaj.art-2fd432a1837b4a2e8068c53feed20d66 |
| institution | Directory Open Access Journal |
| issn | 1551-0018 |
| language | English |
| last_indexed | 2024-03-08T00:02:48Z |
| publishDate | 2024-01-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Mathematical Biosciences and Engineering |
| spelling | doaj.art-2fd432a1837b4a2e8068c53feed20d662024-02-18T01:39:11ZengAIMS PressMathematical Biosciences and Engineering1551-00182024-01-012122432245710.3934/mbe.2024107An adaptive multitasking optimization algorithm based on population distributionXiaoyu Li0Lei Wang1Qiaoyong Jiang2Qingzheng Xu31. School of Computer Science and Engineering, Xi'an University of Technology, Xi'an 710048, China 2. School of Electronic and Information Engineering, Ankang University, Ankang 725000, China1. School of Computer Science and Engineering, Xi'an University of Technology, Xi'an 710048, China3. The Key Laboratory of Industrial Automation of Shaanxi Province, Shaanxi University of Technology, Hanzhong 723001, China1. School of Computer Science and Engineering, Xi'an University of Technology, Xi'an 710048, China4. College of Information and Communication, National University of Defense Technology, Wuhan 430019, ChinaEvolutionary multitasking optimization (EMTO) handles multiple tasks simultaneously by transferring and sharing valuable knowledge from other relevant tasks. How to effectively identify transferred knowledge and reduce negative knowledge transfer are two key issues in EMTO. Many existing EMTO algorithms treat the elite solutions in tasks as transferred knowledge between tasks. However, these algorithms may not be effective enough when the global optimums of the tasks are far apart. In this paper, we study an adaptive evolutionary multitasking optimization algorithm based on population distribution information to find valuable transferred knowledge and weaken the negative transfer between tasks. In this paper, we first divide each task population into K sub-populations based on the fitness values of the individuals, and then the maximum mean discrepancy (MMD) is utilized to calculate the distribution difference between each sub-population in the source task and the sub-population where the best solution of the target task is located. Among the sub-populations of the source task, the sub-population with the smallest MMD value is selected, and the individuals in it are used as transferred individuals. In this way, the solution chosen for the transfer may be an elite solution or some other solution. In addition, an improved randomized interaction probability is also included in the proposed algorithm to adjust the intensity of inter-task interactions. The experimental results on two multitasking test suites demonstrate that the proposed algorithm achieves high solution accuracy and fast convergence for most problems, especially for problems with low relevance.https://www.aimspress.com/article/doi/10.3934/mbe.2024107?viewType=HTMLevolutionary multitasking optimizationdifferential evolutionpopulation distribution informationmaximum mean differencesimilarity |
| spellingShingle | Xiaoyu Li Lei Wang Qiaoyong Jiang Qingzheng Xu An adaptive multitasking optimization algorithm based on population distribution Mathematical Biosciences and Engineering evolutionary multitasking optimization differential evolution population distribution information maximum mean difference similarity |
| title | An adaptive multitasking optimization algorithm based on population distribution |
| title_full | An adaptive multitasking optimization algorithm based on population distribution |
| title_fullStr | An adaptive multitasking optimization algorithm based on population distribution |
| title_full_unstemmed | An adaptive multitasking optimization algorithm based on population distribution |
| title_short | An adaptive multitasking optimization algorithm based on population distribution |
| title_sort | adaptive multitasking optimization algorithm based on population distribution |
| topic | evolutionary multitasking optimization differential evolution population distribution information maximum mean difference similarity |
| url | https://www.aimspress.com/article/doi/10.3934/mbe.2024107?viewType=HTML |
| work_keys_str_mv | AT xiaoyuli anadaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT leiwang anadaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT qiaoyongjiang anadaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT qingzhengxu anadaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT xiaoyuli adaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT leiwang adaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT qiaoyongjiang adaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution AT qingzhengxu adaptivemultitaskingoptimizationalgorithmbasedonpopulationdistribution |