Transfer Learning for Operator Selection: A Reinforcement Learning Approach
In the past two decades, metaheuristic optimisation algorithms (MOAs) have been increasingly popular, particularly in logistic, science, and engineering problems. The fundamental characteristics of such algorithms are that they are dependent on a parameter or a strategy. Some online and offline stra...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/15/1/24 |
| _version_ | 1797496430694236160 |
|---|---|
| author | Rafet Durgut Mehmet Emin Aydin Abdur Rakib |
| author_facet | Rafet Durgut Mehmet Emin Aydin Abdur Rakib |
| author_sort | Rafet Durgut |
| collection | DOAJ |
| description | In the past two decades, metaheuristic optimisation algorithms (MOAs) have been increasingly popular, particularly in logistic, science, and engineering problems. The fundamental characteristics of such algorithms are that they are dependent on a parameter or a strategy. Some online and offline strategies are employed in order to obtain optimal configurations of the algorithms. Adaptive operator selection is one of them, and it determines whether or not to update a strategy from the strategy pool during the search process. In the field of machine learning, Reinforcement Learning (RL) refers to goal-oriented algorithms, which learn from the environment how to achieve a goal. On MOAs, reinforcement learning has been utilised to control the operator selection process. However, existing research fails to show that learned information may be transferred from one problem-solving procedure to another. The primary goal of the proposed research is to determine the impact of transfer learning on RL and MOAs. As a test problem, a set union knapsack problem with 30 separate benchmark problem instances is used. The results are statistically compared in depth. The learning process, according to the findings, improved the convergence speed while significantly reducing the CPU time. |
| first_indexed | 2024-03-10T03:04:28Z |
| format | Article |
| id | doaj.art-3608da497c84436c9f2e6ff6b99082ef |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-03-10T03:04:28Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-3608da497c84436c9f2e6ff6b99082ef2023-11-23T12:40:48ZengMDPI AGAlgorithms1999-48932022-01-011512410.3390/a15010024Transfer Learning for Operator Selection: A Reinforcement Learning ApproachRafet Durgut0Mehmet Emin Aydin1Abdur Rakib2Department of Computer Engineering, Bandirma Onyedi Eylul University, Bandirma 10200, TurkeyDepartment of Computer Science and Creative Technologies, UWE Bristol, Bristol BS16 1QY, UKDepartment of Computer Science and Creative Technologies, UWE Bristol, Bristol BS16 1QY, UKIn the past two decades, metaheuristic optimisation algorithms (MOAs) have been increasingly popular, particularly in logistic, science, and engineering problems. The fundamental characteristics of such algorithms are that they are dependent on a parameter or a strategy. Some online and offline strategies are employed in order to obtain optimal configurations of the algorithms. Adaptive operator selection is one of them, and it determines whether or not to update a strategy from the strategy pool during the search process. In the field of machine learning, Reinforcement Learning (RL) refers to goal-oriented algorithms, which learn from the environment how to achieve a goal. On MOAs, reinforcement learning has been utilised to control the operator selection process. However, existing research fails to show that learned information may be transferred from one problem-solving procedure to another. The primary goal of the proposed research is to determine the impact of transfer learning on RL and MOAs. As a test problem, a set union knapsack problem with 30 separate benchmark problem instances is used. The results are statistically compared in depth. The learning process, according to the findings, improved the convergence speed while significantly reducing the CPU time.https://www.mdpi.com/1999-4893/15/1/24transfer learningreinforcement learningadaptive operator selectionartificial bee colony |
| spellingShingle | Rafet Durgut Mehmet Emin Aydin Abdur Rakib Transfer Learning for Operator Selection: A Reinforcement Learning Approach Algorithms transfer learning reinforcement learning adaptive operator selection artificial bee colony |
| title | Transfer Learning for Operator Selection: A Reinforcement Learning Approach |
| title_full | Transfer Learning for Operator Selection: A Reinforcement Learning Approach |
| title_fullStr | Transfer Learning for Operator Selection: A Reinforcement Learning Approach |
| title_full_unstemmed | Transfer Learning for Operator Selection: A Reinforcement Learning Approach |
| title_short | Transfer Learning for Operator Selection: A Reinforcement Learning Approach |
| title_sort | transfer learning for operator selection a reinforcement learning approach |
| topic | transfer learning reinforcement learning adaptive operator selection artificial bee colony |
| url | https://www.mdpi.com/1999-4893/15/1/24 |
| work_keys_str_mv | AT rafetdurgut transferlearningforoperatorselectionareinforcementlearningapproach AT mehmeteminaydin transferlearningforoperatorselectionareinforcementlearningapproach AT abdurrakib transferlearningforoperatorselectionareinforcementlearningapproach |