Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution
Extreme learning machine (ELM) is massively mapped to the saturation region of the activation function. Moreover, the input and output of the hidden layer are far from being able to obtain a common distribution method, which gives rise to poor generalization performance. Aiming at this problem, the...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Journal of Computer Engineering and Applications Beijing Co., Ltd., Science Press
2021-04-01
|
| Series: | Jisuanji kexue yu tansuo |
| Subjects: | |
| Online Access: | http://fcst.ceaj.org/CN/abstract/abstract2654.shtml |
| _version_ | 1818887520886194176 |
|---|---|
| author | ZHANG Yi, WANG Shitong |
| author_facet | ZHANG Yi, WANG Shitong |
| author_sort | ZHANG Yi, WANG Shitong |
| collection | DOAJ |
| description | Extreme learning machine (ELM) is massively mapped to the saturation region of the activation function. Moreover, the input and output of the hidden layer are far from being able to obtain a common distribution method, which gives rise to poor generalization performance. Aiming at this problem, the extreme learning machine that optimizes the affine transformation (AT) in the activation function under the Gaussian distribution is studied. The proposed algorithm introduces a new linear relationship of input data in the hidden layer. The gradient descent algorithm is used to optimize the scaling parameters and translation parameters in the objective function to satisfy the hidden layer output highly obeying the Gaussian distribution. The new method of calculating affine parameters based on the Gaussian distribution can ensure that the hidden nodes are independent of each other while retaining a high degree of dependency. The experimental results show that the output data of the hidden layer do not obey the uniform distribution well in the actual classification dataset and the image regression dataset, but obey the Gaussian distribution trend, which can achieve better experimental results in general. Compared with the original ELM algorithm and the AT-ELM1 algorithm, there are significant improvements in general. |
| first_indexed | 2024-12-19T16:38:33Z |
| format | Article |
| id | doaj.art-6ebf21c1d6044bd9be1f37eefc67ee54 |
| institution | Directory Open Access Journal |
| issn | 1673-9418 |
| language | zho |
| last_indexed | 2024-12-19T16:38:33Z |
| publishDate | 2021-04-01 |
| publisher | Journal of Computer Engineering and Applications Beijing Co., Ltd., Science Press |
| record_format | Article |
| series | Jisuanji kexue yu tansuo |
| spelling | doaj.art-6ebf21c1d6044bd9be1f37eefc67ee542022-12-21T20:13:53ZzhoJournal of Computer Engineering and Applications Beijing Co., Ltd., Science PressJisuanji kexue yu tansuo1673-94182021-04-0115469070110.3778/j.issn.1673-9418.1912074Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian DistributionZHANG Yi, WANG Shitong01. School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu 214122, China 2. Key Laboratory of Media Design and Software Technology of Jiangsu Province, Jiangnan University, Wuxi, Jiangsu 214122, ChinaExtreme learning machine (ELM) is massively mapped to the saturation region of the activation function. Moreover, the input and output of the hidden layer are far from being able to obtain a common distribution method, which gives rise to poor generalization performance. Aiming at this problem, the extreme learning machine that optimizes the affine transformation (AT) in the activation function under the Gaussian distribution is studied. The proposed algorithm introduces a new linear relationship of input data in the hidden layer. The gradient descent algorithm is used to optimize the scaling parameters and translation parameters in the objective function to satisfy the hidden layer output highly obeying the Gaussian distribution. The new method of calculating affine parameters based on the Gaussian distribution can ensure that the hidden nodes are independent of each other while retaining a high degree of dependency. The experimental results show that the output data of the hidden layer do not obey the uniform distribution well in the actual classification dataset and the image regression dataset, but obey the Gaussian distribution trend, which can achieve better experimental results in general. Compared with the original ELM algorithm and the AT-ELM1 algorithm, there are significant improvements in general.http://fcst.ceaj.org/CN/abstract/abstract2654.shtmlextreme learning machine (elm)affine transformation (at)gaussian distribution; classification |
| spellingShingle | ZHANG Yi, WANG Shitong Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution Jisuanji kexue yu tansuo extreme learning machine (elm) affine transformation (at) gaussian distribution; classification |
| title | Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution |
| title_full | Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution |
| title_fullStr | Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution |
| title_full_unstemmed | Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution |
| title_short | Extreme Learning Machine for Optimized Affine Transformation Based on Gaussian Distribution |
| title_sort | extreme learning machine for optimized affine transformation based on gaussian distribution |
| topic | extreme learning machine (elm) affine transformation (at) gaussian distribution; classification |
| url | http://fcst.ceaj.org/CN/abstract/abstract2654.shtml |
| work_keys_str_mv | AT zhangyiwangshitong extremelearningmachineforoptimizedaffinetransformationbasedongaussiandistribution |