Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree
Abstract Multifactorial optimization (MFO) is a kind of optimization problem that has attracted considerable attention in recent years. The multifactorial evolutionary algorithm utilizes the implicit genetic transfer mechanism characterized by knowledge transfer to conduct evolutionary multitasking...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-05-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-023-01105-4 |
| _version_ | 1797647157916860416 |
|---|---|
| author | Wei Li Xinyu Gao Lei Wang |
| author_facet | Wei Li Xinyu Gao Lei Wang |
| author_sort | Wei Li |
| collection | DOAJ |
| description | Abstract Multifactorial optimization (MFO) is a kind of optimization problem that has attracted considerable attention in recent years. The multifactorial evolutionary algorithm utilizes the implicit genetic transfer mechanism characterized by knowledge transfer to conduct evolutionary multitasking simultaneously. Therefore, the effectiveness of knowledge transfer significantly affects the performance of the algorithm. To achieve positive knowledge transfer, this paper proposed an evolutionary multitasking optimization algorithm with adaptive transfer strategy based on the decision tree (EMT-ADT). To evaluate the useful knowledge contained in the transferred individuals, this paper defines an evaluation indicator to quantify the transfer ability of each individual. Furthermore, a decision tree is constructed to predict the transfer ability of transferred individuals. Based on the prediction results, promising positive-transferred individuals are selected to transfer knowledge, which can effectively improve the performance of the algorithm. Finally, CEC2017 MFO benchmark problems, WCCI20-MTSO and WCCI20-MaTSO benchmark problems are used to verify the performance of the proposed algorithm EMT-ADT. Experimental results demonstrate the competiveness of EMT-ADT compared with some state-of-the-art algorithms. |
| first_indexed | 2024-03-11T15:12:18Z |
| format | Article |
| id | doaj.art-fcb255135e9543b6bf2c34ea4cbbe35a |
| institution | Directory Open Access Journal |
| issn | 2199-4536 2198-6053 |
| language | English |
| last_indexed | 2024-03-11T15:12:18Z |
| publishDate | 2023-05-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj.art-fcb255135e9543b6bf2c34ea4cbbe35a2023-10-29T12:41:23ZengSpringerComplex & Intelligent Systems2199-45362198-60532023-05-01966697672810.1007/s40747-023-01105-4Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision treeWei Li0Xinyu Gao1Lei Wang2School of Computer Science and Engineering, Xi’an University of TechnologySchool of Computer Science and Engineering, Xi’an University of TechnologyShaanxi Key Laboratory for Network Computing and Security TechnologyAbstract Multifactorial optimization (MFO) is a kind of optimization problem that has attracted considerable attention in recent years. The multifactorial evolutionary algorithm utilizes the implicit genetic transfer mechanism characterized by knowledge transfer to conduct evolutionary multitasking simultaneously. Therefore, the effectiveness of knowledge transfer significantly affects the performance of the algorithm. To achieve positive knowledge transfer, this paper proposed an evolutionary multitasking optimization algorithm with adaptive transfer strategy based on the decision tree (EMT-ADT). To evaluate the useful knowledge contained in the transferred individuals, this paper defines an evaluation indicator to quantify the transfer ability of each individual. Furthermore, a decision tree is constructed to predict the transfer ability of transferred individuals. Based on the prediction results, promising positive-transferred individuals are selected to transfer knowledge, which can effectively improve the performance of the algorithm. Finally, CEC2017 MFO benchmark problems, WCCI20-MTSO and WCCI20-MaTSO benchmark problems are used to verify the performance of the proposed algorithm EMT-ADT. Experimental results demonstrate the competiveness of EMT-ADT compared with some state-of-the-art algorithms.https://doi.org/10.1007/s40747-023-01105-4Multifactorial optimizationEvolutionary algorithmKnowledge transferDecision tree |
| spellingShingle | Wei Li Xinyu Gao Lei Wang Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree Complex & Intelligent Systems Multifactorial optimization Evolutionary algorithm Knowledge transfer Decision tree |
| title | Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| title_full | Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| title_fullStr | Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| title_full_unstemmed | Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| title_short | Multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| title_sort | multifactorial evolutionary algorithm with adaptive transfer strategy based on decision tree |
| topic | Multifactorial optimization Evolutionary algorithm Knowledge transfer Decision tree |
| url | https://doi.org/10.1007/s40747-023-01105-4 |
| work_keys_str_mv | AT weili multifactorialevolutionaryalgorithmwithadaptivetransferstrategybasedondecisiontree AT xinyugao multifactorialevolutionaryalgorithmwithadaptivetransferstrategybasedondecisiontree AT leiwang multifactorialevolutionaryalgorithmwithadaptivetransferstrategybasedondecisiontree |