Vote-boosting ensembles

Vote-boosting is a sequential ensemble learning method in which the individual classi ers are built on diferent weighted versions of the training data. To build a new classi er, the weight of each training instance is determined in terms of the degree of disagreement among the current ensemble predi...

Descripción completa

Detalles Bibliográficos
Autores: Sabzevari, Maryam, Martínez Muñoz, Gonzalo, Suárez González, Alberto
Tipo de recurso: artículo
Fecha de publicación:2018
País:España
Institución:Universidad Autónoma de Madrid
Repositorio:Biblos-e Archivo. Repositorio Institucional de la UAM
Idioma:inglés
OAI Identifier:oai:repositorio.uam.es:10486/692397
Acceso en línea:http://hdl.handle.net/10486/692397
https://dx.doi.org/10.1016/j.patcog.2018.05.022
Access Level:acceso abierto
Palabra clave:Ensemble learning
Boosting
Uncertainty-based emphasis
Robust classification
Informática
Descripción
Sumario:Vote-boosting is a sequential ensemble learning method in which the individual classi ers are built on diferent weighted versions of the training data. To build a new classi er, the weight of each training instance is determined in terms of the degree of disagreement among the current ensemble predictions for that instance. For low class-label noise levels, especially when simple base learners are used, emphasis should be made on instances for which the disagreement rate is high. When more exible classi ers are used and as the noise level increases, the emphasis on these uncertain instances should be reduced. In fact, at suffciently high levels of class-label noise, the focus should be on instances on which the ensemble classi ers agree. The optimal type of emphasis can be automatically determined using cross-validation. An extensive empirical analysis using the beta distribution as emphasis function illustrates that vote-boosting is an effective method to generate ensembles that are both accurate and robust