A Filter-SQP strategy for training Support Vector Machine models
This paper introduces a filtering strategy for addressing optimization problems arising in binary Support Vector Machine classification. The training optimization problem aims to solve the dual formulation which involves a quadratic objective function subjected to a linear and box constraints. Our...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Instituto Federal de Educação, Ciência e Tecnologia do Rio Grande do Sul (IFRS)
2023-10-01
|
| Series: | REMAT |
| Subjects: | |
| Online Access: | https://periodicos.ifrs.edu.br/index.php/REMAT/article/view/6241 |
| _version_ | 1797644670850826240 |
|---|---|
| author | Tiago Lino Bello Luiz Carlos Matioli Lucas Garcia Pedroso Daniela Miray Igarashi |
| author_facet | Tiago Lino Bello Luiz Carlos Matioli Lucas Garcia Pedroso Daniela Miray Igarashi |
| author_sort | Tiago Lino Bello |
| collection | DOAJ |
| description |
This paper introduces a filtering strategy for addressing optimization problems arising in binary Support Vector Machine classification. The training optimization problem aims to solve the dual formulation which involves a quadratic objective function subjected to a linear and box constraints. Our approach employs a Filter algorithm with Sequential Quadratic Programming iterations that minimize the quadratic Lagrangian approximations. Notably, we utilize the exact Hessian matrix in our numerical experiments to seek the desired classification function. Moreover, we present a Filter algorithm combined with the Augmented Lagrangian method aiming to accelerate the algorithm convergence. To substantiate our method's effectiveness, we conduct numerical experiments through MATLAB, comparing outcomes with alternative methodologies detailed in existing literature. Numerical experiments shows that the Filter--SQP combined with Augmented Lagrangian method is competitive and efficient method compared with an interior-point based solver and LIBSVM software in relation of classification metrics and CPU-time.
|
| first_indexed | 2024-03-11T14:34:01Z |
| format | Article |
| id | doaj.art-1f57f34ea3b542548d4e51558444551f |
| institution | Directory Open Access Journal |
| issn | 2447-2689 |
| language | English |
| last_indexed | 2024-03-11T14:34:01Z |
| publishDate | 2023-10-01 |
| publisher | Instituto Federal de Educação, Ciência e Tecnologia do Rio Grande do Sul (IFRS) |
| record_format | Article |
| series | REMAT |
| spelling | doaj.art-1f57f34ea3b542548d4e51558444551f2023-10-31T04:37:20ZengInstituto Federal de Educação, Ciência e Tecnologia do Rio Grande do Sul (IFRS)REMAT2447-26892023-10-019210.35819/remat2023v9i2id6241A Filter-SQP strategy for training Support Vector Machine modelsTiago Lino Bello0Luiz Carlos Matioli1Lucas Garcia Pedroso2Daniela Miray Igarashi3Universidade Federal do Paraná (UFPR), Curitiba, PR, BrazilFederal University of Paraná (UFPR), Curitiba, PR, BrazilFederal University of Paraná (UFPR), Curitiba, PR, BrazilFederal University of Paraná (UFPR), Curitiba, PR, Brazil This paper introduces a filtering strategy for addressing optimization problems arising in binary Support Vector Machine classification. The training optimization problem aims to solve the dual formulation which involves a quadratic objective function subjected to a linear and box constraints. Our approach employs a Filter algorithm with Sequential Quadratic Programming iterations that minimize the quadratic Lagrangian approximations. Notably, we utilize the exact Hessian matrix in our numerical experiments to seek the desired classification function. Moreover, we present a Filter algorithm combined with the Augmented Lagrangian method aiming to accelerate the algorithm convergence. To substantiate our method's effectiveness, we conduct numerical experiments through MATLAB, comparing outcomes with alternative methodologies detailed in existing literature. Numerical experiments shows that the Filter--SQP combined with Augmented Lagrangian method is competitive and efficient method compared with an interior-point based solver and LIBSVM software in relation of classification metrics and CPU-time. https://periodicos.ifrs.edu.br/index.php/REMAT/article/view/6241support vector machinetrainingoptimizationfilter methodsequential quadratic programming |
| spellingShingle | Tiago Lino Bello Luiz Carlos Matioli Lucas Garcia Pedroso Daniela Miray Igarashi A Filter-SQP strategy for training Support Vector Machine models REMAT support vector machine training optimization filter method sequential quadratic programming |
| title | A Filter-SQP strategy for training Support Vector Machine models |
| title_full | A Filter-SQP strategy for training Support Vector Machine models |
| title_fullStr | A Filter-SQP strategy for training Support Vector Machine models |
| title_full_unstemmed | A Filter-SQP strategy for training Support Vector Machine models |
| title_short | A Filter-SQP strategy for training Support Vector Machine models |
| title_sort | filter sqp strategy for training support vector machine models |
| topic | support vector machine training optimization filter method sequential quadratic programming |
| url | https://periodicos.ifrs.edu.br/index.php/REMAT/article/view/6241 |
| work_keys_str_mv | AT tiagolinobello afiltersqpstrategyfortrainingsupportvectormachinemodels AT luizcarlosmatioli afiltersqpstrategyfortrainingsupportvectormachinemodels AT lucasgarciapedroso afiltersqpstrategyfortrainingsupportvectormachinemodels AT danielamirayigarashi afiltersqpstrategyfortrainingsupportvectormachinemodels AT tiagolinobello filtersqpstrategyfortrainingsupportvectormachinemodels AT luizcarlosmatioli filtersqpstrategyfortrainingsupportvectormachinemodels AT lucasgarciapedroso filtersqpstrategyfortrainingsupportvectormachinemodels AT danielamirayigarashi filtersqpstrategyfortrainingsupportvectormachinemodels |