On the Consistency of Ordinal Regression Methods

Fabian Pedregosa 1, 2 Francis Bach 1, 2 Alexandre Gramfort 3
2 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, CNRS - Centre National de la Recherche Scientifique, Inria de Paris
3 PARIETAL - Modelling brain structure, function and variability based on high-field MRI data
NEUROSPIN - Service NEUROSPIN, Inria Saclay - Ile de France
Abstract : 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.
Document type :
Journal articles
Complete list of metadatas

https://hal.inria.fr/hal-01054942
Contributor : Fabian Pedregosa <>
Submitted on : Monday, June 19, 2017 - 8:29:14 PM
Last modification on : Friday, March 8, 2019 - 1:20:24 AM
Long-term archiving on : Friday, December 15, 2017 - 8:17:00 PM

Files

15-495.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

  • HAL Id : hal-01054942, version 4
  • ARXIV : 1408.2327

Citation

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

Share

Metrics

Record views

982

Files downloads

250