A Generalized Lagrange Multiplier Method Support for Vector Regression Based

This research presents an approach to support vector regression based on the epsilon L1 and L2 formulations. In contrast to standard architectures, it explores a new formulation where the dual optimization problem results from formulating an extended Lagrangian function, introducing additional terms...

Descripción completa

Detalles Bibliográficos
Autor: Rodríguez-Reyes, Sara E.
Tipo de recurso: tesis de maestría
Estado:Versión aceptada para publicación
Fecha de publicación:2021
País:México
Institución:Instituto Tecnológico y de Estudios Superiores de Occidente
Repositorio:Repositorio Institucional del ITESO
Idioma:inglés
OAI Identifier:oai:rei.iteso.mx:11117/7434
Acceso en línea:https://hdl.handle.net/11117/7434
Access Level:acceso abierto
Palabra clave:Extended Lagrangian
Kernel-Based Methods
Support Vector Regression
Descripción
Sumario:This research presents an approach to support vector regression based on the epsilon L1 and L2 formulations. In contrast to standard architectures, it explores a new formulation where the dual optimization problem results from formulating an extended Lagrangian function, introducing additional terms to include a weighted elastic net regularization structure. Additionally, the research shows the differences and similarities of this proposal with the classical support vector regression and the LASSO regression, aiming to compare them with standard models. To demonstrate the capabilities of this approach, the document includes examples of predicting some benchmark functions.