Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information
Abstract In ensemble learning, random subspace technology not only easily loses some important features but also easily produces some redundant subspaces, inevitably leading to the decline of ensemble learning performance. In order to overcome the shortcomings, we propose a new selective quantum ens...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-03-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-023-00996-7 |
| _version_ | 1797674960088465408 |
|---|---|
| author | Xufeng Niu Wenping Ma |
| author_facet | Xufeng Niu Wenping Ma |
| author_sort | Xufeng Niu |
| collection | DOAJ |
| description | Abstract In ensemble learning, random subspace technology not only easily loses some important features but also easily produces some redundant subspaces, inevitably leading to the decline of ensemble learning performance. In order to overcome the shortcomings, we propose a new selective quantum ensemble learning model inspired by improved AdaBoost based on local sample information (SELA). Firstly, SELA combines information entropy and random subspace to ensure that the important features of the classification task in each subspace are preserved. Then, we select the base classifier that can balance accuracy and diversity among a group of base classifiers generated based on local AdaBoost in each iteration. Finally, we utilize the quantum genetic algorithm to search optimal weights for base learners in the label prediction process. We use UCI datasets to analyze the impact of important parameters in SELA on classification performance and verify that SELA is usually superior to other competitive algorithms. |
| first_indexed | 2024-03-11T22:07:45Z |
| format | Article |
| id | doaj.art-3ecbe273b8ea407fbe6ff5dfb839bd7f |
| institution | Directory Open Access Journal |
| issn | 2199-4536 2198-6053 |
| language | English |
| last_indexed | 2024-03-11T22:07:45Z |
| publishDate | 2023-03-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj.art-3ecbe273b8ea407fbe6ff5dfb839bd7f2023-09-24T11:35:42ZengSpringerComplex & Intelligent Systems2199-45362198-60532023-03-01955173518310.1007/s40747-023-00996-7Selective quantum ensemble learning inspired by improved AdaBoost based on local sample informationXufeng Niu0Wenping Ma1School of Telecommunication Engineering, Xidian UniversitySchool of Telecommunication Engineering, Xidian UniversityAbstract In ensemble learning, random subspace technology not only easily loses some important features but also easily produces some redundant subspaces, inevitably leading to the decline of ensemble learning performance. In order to overcome the shortcomings, we propose a new selective quantum ensemble learning model inspired by improved AdaBoost based on local sample information (SELA). Firstly, SELA combines information entropy and random subspace to ensure that the important features of the classification task in each subspace are preserved. Then, we select the base classifier that can balance accuracy and diversity among a group of base classifiers generated based on local AdaBoost in each iteration. Finally, we utilize the quantum genetic algorithm to search optimal weights for base learners in the label prediction process. We use UCI datasets to analyze the impact of important parameters in SELA on classification performance and verify that SELA is usually superior to other competitive algorithms.https://doi.org/10.1007/s40747-023-00996-7Selective ensemble learningQuantum computingGenetic algorithmLocal AdaBoost |
| spellingShingle | Xufeng Niu Wenping Ma Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information Complex & Intelligent Systems Selective ensemble learning Quantum computing Genetic algorithm Local AdaBoost |
| title | Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information |
| title_full | Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information |
| title_fullStr | Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information |
| title_full_unstemmed | Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information |
| title_short | Selective quantum ensemble learning inspired by improved AdaBoost based on local sample information |
| title_sort | selective quantum ensemble learning inspired by improved adaboost based on local sample information |
| topic | Selective ensemble learning Quantum computing Genetic algorithm Local AdaBoost |
| url | https://doi.org/10.1007/s40747-023-00996-7 |
| work_keys_str_mv | AT xufengniu selectivequantumensemblelearninginspiredbyimprovedadaboostbasedonlocalsampleinformation AT wenpingma selectivequantumensemblelearninginspiredbyimprovedadaboostbasedonlocalsampleinformation |