Kernel Learning by Spectral Representation and Gaussian Mixtures
One of the main tasks in kernel methods is the selection of adequate mappings into higher dimension in order to improve class classification. However, this tends to be time consuming, and it may not finish with the best separation between classes. Therefore, there is a need for better methods that a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/4/2473 |
| _version_ | 1827758816574308352 |
|---|---|
| author | Luis R. Pena-Llamas Ramon O. Guardado-Medina Arturo Garcia Andres Mendez-Vazquez |
| author_facet | Luis R. Pena-Llamas Ramon O. Guardado-Medina Arturo Garcia Andres Mendez-Vazquez |
| author_sort | Luis R. Pena-Llamas |
| collection | DOAJ |
| description | One of the main tasks in kernel methods is the selection of adequate mappings into higher dimension in order to improve class classification. However, this tends to be time consuming, and it may not finish with the best separation between classes. Therefore, there is a need for better methods that are able to extract distance and class separation from data. This work presents a novel approach for learning such mappings by using locally stationary kernels, spectral representations and Gaussian mixtures. |
| first_indexed | 2024-03-11T09:11:50Z |
| format | Article |
| id | doaj.art-3d2ae2bef9104fc98b377b7f053dd69a |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T09:11:50Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-3d2ae2bef9104fc98b377b7f053dd69a2023-11-16T18:56:14ZengMDPI AGApplied Sciences2076-34172023-02-01134247310.3390/app13042473Kernel Learning by Spectral Representation and Gaussian MixturesLuis R. Pena-Llamas0Ramon O. Guardado-Medina1Arturo Garcia2Andres Mendez-Vazquez3Department of Computer Science, El Centro de Investigación y de Estudios Avanzados (CINVESTAV), Ciudad de Mexico 44960, MexicoDepartment of Research, Escuela Militar de Mantenimiento y Abastecimiento, Universidad del Ejercito y Fuerza Aerea, Zapopan 45200, MexicoDepartment of Research, Escuela Militar de Mantenimiento y Abastecimiento, Universidad del Ejercito y Fuerza Aerea, Zapopan 45200, MexicoDepartment of Computer Science, El Centro de Investigación y de Estudios Avanzados (CINVESTAV), Ciudad de Mexico 44960, MexicoOne of the main tasks in kernel methods is the selection of adequate mappings into higher dimension in order to improve class classification. However, this tends to be time consuming, and it may not finish with the best separation between classes. Therefore, there is a need for better methods that are able to extract distance and class separation from data. This work presents a novel approach for learning such mappings by using locally stationary kernels, spectral representations and Gaussian mixtures.https://www.mdpi.com/2076-3417/13/4/2473non-parametric kernel learningapproximating kernelkernel spectral representationlocally stationary kernel |
| spellingShingle | Luis R. Pena-Llamas Ramon O. Guardado-Medina Arturo Garcia Andres Mendez-Vazquez Kernel Learning by Spectral Representation and Gaussian Mixtures Applied Sciences non-parametric kernel learning approximating kernel kernel spectral representation locally stationary kernel |
| title | Kernel Learning by Spectral Representation and Gaussian Mixtures |
| title_full | Kernel Learning by Spectral Representation and Gaussian Mixtures |
| title_fullStr | Kernel Learning by Spectral Representation and Gaussian Mixtures |
| title_full_unstemmed | Kernel Learning by Spectral Representation and Gaussian Mixtures |
| title_short | Kernel Learning by Spectral Representation and Gaussian Mixtures |
| title_sort | kernel learning by spectral representation and gaussian mixtures |
| topic | non-parametric kernel learning approximating kernel kernel spectral representation locally stationary kernel |
| url | https://www.mdpi.com/2076-3417/13/4/2473 |
| work_keys_str_mv | AT luisrpenallamas kernellearningbyspectralrepresentationandgaussianmixtures AT ramonoguardadomedina kernellearningbyspectralrepresentationandgaussianmixtures AT arturogarcia kernellearningbyspectralrepresentationandgaussianmixtures AT andresmendezvazquez kernellearningbyspectralrepresentationandgaussianmixtures |