On the Consistency of Ordinal Regression Methods - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Machine Learning Research Année : 2017

On the Consistency of Ordinal Regression Methods

Résumé

Many of the ordinal regression models that have been proposed in the literature can be seen as methods that minimize a convex surrogate of the zero-one, absolute, or squared loss functions. A key property that allows to study the statistical implications of such approximations is that of Fisher consistency. Fisher consistency is a desirable property for surrogate loss functions and implies that in the population setting, i.e., if the probability distribution that generates the data were available, then optimization of the surrogate would yield the best possible model. In this paper we will characterize the Fisher consistency of a rich family of surrogate loss functions used in the context of ordinal regression, including support vector ordinal regression, ORBoosting and least absolute deviation. We will see that, for a family of surrogate loss functions that subsumes support vector ordinal regression and ORBoosting, consistency can be fully characterized by the derivative of a real-valued function at zero, as happens for convex margin-based surrogates in binary classification. We also derive excess risk bounds for a surrogate of the absolute error that generalize existing risk bounds for binary classification. Finally, our analysis suggests a novel surrogate of the squared error loss. We compare this novel surrogate with competing approaches on 9 different datasets. Our method shows to be highly competitive in practice, outperforming the least squares loss on 7 out of 9 datasets.
Fichier principal
Vignette du fichier
15-495.pdf (613.41 Ko) Télécharger le fichier
counterexample.pdf (92.29 Ko) Télécharger le fichier
scores.pdf (176.25 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01054942 , version 1 (10-08-2014)
hal-01054942 , version 2 (27-10-2014)
hal-01054942 , version 3 (29-09-2015)
hal-01054942 , version 4 (19-06-2017)

Licence

Paternité

Identifiants

Citer

Fabian Pedregosa, Francis Bach, Alexandre Gramfort. On the Consistency of Ordinal Regression Methods. Journal of Machine Learning Research, 2017, 18, pp.1 - 35. ⟨hal-01054942v4⟩
1249 Consultations
997 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More