Ordered Weighted Average Support Vector Regression
Abstract
This paper introduces a novel Support Vector Regression (SVR) model that incorporates Ordered Weighted Average (OWA) operators to differently penalize deviations of observations from the ε-strip. The penalty is determined based on the position of each deviation in the ordered vector of all deviations. A key contribution of this work is the development of two nonlinear formulations: a continuous formulation for non-decreasing monotone weight vectors and a mixed-integer formulation for general weight vectors. By leveraging dual approaches associated with these formulations, the model accommodates nonlinear kernel functions. To enhance computational efficiency, two heuristic approaches are proposed for deriving suitable regression hyperplanes in reduced computation times, as compared to exact methods. Additionally, a third heuristic approach is developed to handle nonlinear kernel functions effectively. Computational experiments conducted on both real and synthetic datasets demonstrate that the exact formulations yield remarkable regression functions with respect to the following standard metrics: mean absolute errors (MAE) and mean squared errors (MSE). The heuristic approaches are shown to be particularly efficient for larger datasets, striking a balance between solution quality and computational effort.
Versions
Related
- Support Vector Machines for Classification, Regression, and Outlier Detection: A Tutorial
- Exact Recovery Thresholds for Weighted Data Selection in Vector-Valued Linear Regression
- On stable semilinear sets in ordered vector spaces
- Order convergence in ordered vector spaces: compactness, atoms, pervasiveness, and ice cream spaces
- Fourth-Order Energy-Consistent Average Vector Field Discrete Gradient Methods for Dissipative Hamiltonian Systems
BibTeX
@misc{martinezmerino2026ordered,
title = {{Ordered Weighted Average Support Vector Regression}},
author = {Luisa I. Martínez-Merino and Justo Puerto and Antonio M. Rodríguez-Chía},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05156},
eprint = {2610.05156},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05156},
}
RIS
TY - UNPB
TI - Ordered Weighted Average Support Vector Regression
AU - Luisa I. Martínez-Merino
AU - Justo Puerto
AU - Antonio M. Rodríguez-Chía
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05156
UR - https://arxiv.org/abs/2610.05156
AB - This paper introduces a novel Support Vector Regression (SVR) model that incorporates Ordered Weighted Average (OWA) operators to differently penalize deviations of observations from the ε-strip. The penalty is determined based on the position of each deviation in the ordered vector of all deviations. A key contribution of this work is the development of two nonlinear formulations: a continuous formulation for non-decreasing monotone weight vectors and a mixed-integer formulation for general weight vectors. By leveraging dual approaches associated with these formulations, the model accommodates nonlinear kernel functions. To enhance computational efficiency, two heuristic approaches are proposed for deriving suitable regression hyperplanes in reduced computation times, as compared to exact methods. Additionally, a third heuristic approach is developed to handle nonlinear kernel functions effectively. Computational experiments conducted on both real and synthetic datasets demonstrate that the exact formulations yield remarkable regression functions with respect to the following standard metrics: mean absolute errors (MAE) and mean squared errors (MSE). The heuristic approaches are shown to be particularly efficient for larger datasets, striking a balance between solution quality and computational effort.
ER -