A Generalized Lagrange Multiplier Method for Support Vector Regression with Imposed Symmetry

This thesis presents an approach to support vector regression that extends the classic Vapnik’s formulation. After recalling that the classic formulation contains a Lasso regularization structure in its dual form, we propose a generalized Lagrangian function with additional terms to include the Ridg...

Descripción completa

Detalles Bibliográficos
Autor: Guerrero-Montaño, Luis A.
Tipo de recurso: tesis de maestría
Estado:Versión aceptada para publicación
Fecha de publicación:2022
País:México
Institución:Instituto Tecnológico y de Estudios Superiores de Occidente
Repositorio:Repositorio Institucional del ITESO
Idioma:inglés
OAI Identifier:oai:rei.iteso.mx:11117/8449
Acceso en línea:https://hdl.handle.net/11117/8449
Access Level:acceso abierto
Palabra clave:SVM
GLMM
SVR
Simetría
Symmetry
Support Vector Machine
Support Vector Regression
Descripción
Sumario:This thesis presents an approach to support vector regression that extends the classic Vapnik’s formulation. After recalling that the classic formulation contains a Lasso regularization structure in its dual form, we propose a generalized Lagrangian function with additional terms to include the Ridge regularization in the dual problem for the case with symmetry. By including both regularization methods, the resulting dual problem with the generalized Lagrangian comprises an elastic net regularization structure. Hence, as an immediate consequence, the classical formulation is a particular case of the current proposal. Finally, to demonstrate the capabilities of this approach, the document includes examples of predicting some benchmark problems.