FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition
Extreme learning machine (ELM) is an interesting algorithm for learning the hidden layer of single layer feed forward neural networks. However, one of the main shortcomings restricting further improvement of ELM is the complexity of singular value decomposition (SVD) for computing the Moore-Penrose...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2017
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/61871/1/A%20fast%20adaptive%20shrinkage.pdf |
| _version_ | 1796977531741536256 |
|---|---|
| author | Mahmood, Saif F. Marhaban, Mohammad Hamiruce Rokhani, Fakhrul Zaman Samsudin, Khairulmizam Arigbabu, Olasimbo Ayodeji |
| author_facet | Mahmood, Saif F. Marhaban, Mohammad Hamiruce Rokhani, Fakhrul Zaman Samsudin, Khairulmizam Arigbabu, Olasimbo Ayodeji |
| author_sort | Mahmood, Saif F. |
| collection | UPM |
| description | Extreme learning machine (ELM) is an interesting algorithm for learning the hidden layer of single layer feed forward neural networks. However, one of the main shortcomings restricting further improvement of ELM is the complexity of singular value decomposition (SVD) for computing the Moore-Penrose generalized inverse of the hidden layer matrix. This paper presents a new algorithm named fast adaptive shrinkage/thresholding algorithm ELM (FASTA-ELM) which uses an extension of forward-backward splitting (FBS) to compute the smallest norm of the output weights in ELM. The proposed FASTA-ELM algorithm is evaluated on face gender recognition problem using 5 benchmarked datasets. The results indicate that FASTA-ELM provides efficient performance and outperforms the standard ELM and two other variants of ELM in terms of generalization ability and computational time. Furthermore, the recognition performance of FASTA-ELM is comparable to other state-of-the-art face gender recognition methods. |
| first_indexed | 2024-03-06T09:41:29Z |
| format | Article |
| id | upm.eprints-61871 |
| institution | Universiti Putra Malaysia |
| language | English |
| last_indexed | 2024-03-06T09:41:29Z |
| publishDate | 2017 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | upm.eprints-618712019-02-25T03:28:36Z http://psasir.upm.edu.my/id/eprint/61871/ FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition Mahmood, Saif F. Marhaban, Mohammad Hamiruce Rokhani, Fakhrul Zaman Samsudin, Khairulmizam Arigbabu, Olasimbo Ayodeji Extreme learning machine (ELM) is an interesting algorithm for learning the hidden layer of single layer feed forward neural networks. However, one of the main shortcomings restricting further improvement of ELM is the complexity of singular value decomposition (SVD) for computing the Moore-Penrose generalized inverse of the hidden layer matrix. This paper presents a new algorithm named fast adaptive shrinkage/thresholding algorithm ELM (FASTA-ELM) which uses an extension of forward-backward splitting (FBS) to compute the smallest norm of the output weights in ELM. The proposed FASTA-ELM algorithm is evaluated on face gender recognition problem using 5 benchmarked datasets. The results indicate that FASTA-ELM provides efficient performance and outperforms the standard ELM and two other variants of ELM in terms of generalization ability and computational time. Furthermore, the recognition performance of FASTA-ELM is comparable to other state-of-the-art face gender recognition methods. Elsevier 2017-01 Article PeerReviewed text en http://psasir.upm.edu.my/id/eprint/61871/1/A%20fast%20adaptive%20shrinkage.pdf Mahmood, Saif F. and Marhaban, Mohammad Hamiruce and Rokhani, Fakhrul Zaman and Samsudin, Khairulmizam and Arigbabu, Olasimbo Ayodeji (2017) FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition. Neurocomputing, 219. 312 - 322. ISSN 0925-2312; ESSN: 1872-8286 https://www.sciencedirect.com/science/article/pii/S0925231216310748 10.1016/j.neucom.2016.09.046 |
| spellingShingle | Mahmood, Saif F. Marhaban, Mohammad Hamiruce Rokhani, Fakhrul Zaman Samsudin, Khairulmizam Arigbabu, Olasimbo Ayodeji FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title | FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title_full | FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title_fullStr | FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title_full_unstemmed | FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title_short | FASTA-ELM: a fast adaptive shrinkage/thresholding algorithm for extreme learning machine and its application to gender recognition |
| title_sort | fasta elm a fast adaptive shrinkage thresholding algorithm for extreme learning machine and its application to gender recognition |
| url | http://psasir.upm.edu.my/id/eprint/61871/1/A%20fast%20adaptive%20shrinkage.pdf |
| work_keys_str_mv | AT mahmoodsaiff fastaelmafastadaptiveshrinkagethresholdingalgorithmforextremelearningmachineanditsapplicationtogenderrecognition AT marhabanmohammadhamiruce fastaelmafastadaptiveshrinkagethresholdingalgorithmforextremelearningmachineanditsapplicationtogenderrecognition AT rokhanifakhrulzaman fastaelmafastadaptiveshrinkagethresholdingalgorithmforextremelearningmachineanditsapplicationtogenderrecognition AT samsudinkhairulmizam fastaelmafastadaptiveshrinkagethresholdingalgorithmforextremelearningmachineanditsapplicationtogenderrecognition AT arigbabuolasimboayodeji fastaelmafastadaptiveshrinkagethresholdingalgorithmforextremelearningmachineanditsapplicationtogenderrecognition |