Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment
With fast learning speed and high accuracy, extreme learning machine (ELM) has achieved great success in pattern recognition and machine learning. Unfortunately, it will fail in the circumstance where plenty of labeled samples for training model are insufficient. The labeled samples are difficult to...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi-Wiley
2023-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2023/5072247 |
| _version_ | 1797871849754853376 |
|---|---|
| author | Shaofei Zang Dongqing Li Chao Ma Jianwei Ma |
| author_facet | Shaofei Zang Dongqing Li Chao Ma Jianwei Ma |
| author_sort | Shaofei Zang |
| collection | DOAJ |
| description | With fast learning speed and high accuracy, extreme learning machine (ELM) has achieved great success in pattern recognition and machine learning. Unfortunately, it will fail in the circumstance where plenty of labeled samples for training model are insufficient. The labeled samples are difficult to obtain due to their high cost. In this paper, we solve this problem with transfer learning and propose joint transfer extreme learning machine (JTELM). First, it applies cross-domain mean approximation (CDMA) to minimize the discrepancy between domains, thus obtaining one ELM model. Second, subspace alignment (sa) and weight approximation are together introduced into the output layer to enhance the capability of knowledge transfer and learn another ELM model. Third, the prediction of test samples is dominated by the two learned ELM models. Finally, a series of experiments are carried out to investigate the performance of JTELM, and the results show that it achieves efficiently the task of transfer learning and performs better than the traditional ELM and other transfer or nontransfer learning methods. |
| first_indexed | 2024-04-10T00:49:32Z |
| format | Article |
| id | doaj.art-374010db3b3440bba4720bcb1967eb8c |
| institution | Directory Open Access Journal |
| issn | 1099-0526 |
| language | English |
| last_indexed | 2024-04-10T00:49:32Z |
| publishDate | 2023-01-01 |
| publisher | Hindawi-Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj.art-374010db3b3440bba4720bcb1967eb8c2023-03-13T11:25:50ZengHindawi-WileyComplexity1099-05262023-01-01202310.1155/2023/5072247Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight AlignmentShaofei Zang0Dongqing Li1Chao Ma2Jianwei Ma3College of Information EngineeringKey Laboratory of Electronics and Information Technology for Space SystemsCollege of Information EngineeringCollege of Information EngineeringWith fast learning speed and high accuracy, extreme learning machine (ELM) has achieved great success in pattern recognition and machine learning. Unfortunately, it will fail in the circumstance where plenty of labeled samples for training model are insufficient. The labeled samples are difficult to obtain due to their high cost. In this paper, we solve this problem with transfer learning and propose joint transfer extreme learning machine (JTELM). First, it applies cross-domain mean approximation (CDMA) to minimize the discrepancy between domains, thus obtaining one ELM model. Second, subspace alignment (sa) and weight approximation are together introduced into the output layer to enhance the capability of knowledge transfer and learn another ELM model. Third, the prediction of test samples is dominated by the two learned ELM models. Finally, a series of experiments are carried out to investigate the performance of JTELM, and the results show that it achieves efficiently the task of transfer learning and performs better than the traditional ELM and other transfer or nontransfer learning methods.http://dx.doi.org/10.1155/2023/5072247 |
| spellingShingle | Shaofei Zang Dongqing Li Chao Ma Jianwei Ma Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment Complexity |
| title | Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment |
| title_full | Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment |
| title_fullStr | Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment |
| title_full_unstemmed | Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment |
| title_short | Joint Transfer Extreme Learning Machine with Cross-Domain Mean Approximation and Output Weight Alignment |
| title_sort | joint transfer extreme learning machine with cross domain mean approximation and output weight alignment |
| url | http://dx.doi.org/10.1155/2023/5072247 |
| work_keys_str_mv | AT shaofeizang jointtransferextremelearningmachinewithcrossdomainmeanapproximationandoutputweightalignment AT dongqingli jointtransferextremelearningmachinewithcrossdomainmeanapproximationandoutputweightalignment AT chaoma jointtransferextremelearningmachinewithcrossdomainmeanapproximationandoutputweightalignment AT jianweima jointtransferextremelearningmachinewithcrossdomainmeanapproximationandoutputweightalignment |