Robust Support Vector Machine With Asymmetric Truncated Generalized Pinball Loss

The support vector machine (SVM) with pinball loss (Pin-SVM) can handle noise sensitivity and instability to re-sampling but loses sparsity. To solve this limitation, SVM with a generalized pinball loss that incorporates an insensitive zone (GP-SVM) was proposed. The GP-SVM can handle sparsity by op...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access Vol. 12; pp. 155696 - 155717
Main Authors: Suppalap, Siwakon, Wangkeeree, Rabian
Format: Journal Article
Language:English
Published: Piscataway IEEE 2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Asymmetry
Classification
Convex functions
difference of convex functions programming
Fasteners
generalized pinball loss
Noise
Noise sensitivity
Optimization methods
pinball loss
Programming
Robustness
Sensitivity
Sparsity
Support vector machine
Support vector machines
Training
Vectors
ISSN:2169-3536, 2169-3536
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The support vector machine (SVM) with pinball loss (Pin-SVM) can handle noise sensitivity and instability to re-sampling but loses sparsity. To solve this limitation, SVM with a generalized pinball loss that incorporates an insensitive zone (GP-SVM) was proposed. The GP-SVM can handle sparsity by optimizing the asymmetric spread of the insensitive zone. Despite these improvements, the unboundedness of the loss functions can result in a lack of robustness to outliers. In this paper, we introduce a novel robust support vector classification based on an <inline-formula> <tex-math notation="LaTeX">(\alpha _{1}, \alpha _{2}) </tex-math></inline-formula>-asymmetric bounded loss function, an asymmetric truncated generalized pinball loss (called <inline-formula> <tex-math notation="LaTeX">L_{tgp}^{\alpha _{1}, \alpha _{2}} </tex-math></inline-formula>). A characteristic of SVM with <inline-formula> <tex-math notation="LaTeX">L_{tgp}^{\alpha _{1}, \alpha _{2}} </tex-math></inline-formula> (ATGP-SVM) is its ability to balance generalization and sparsity while minimizing the impact of outliers. However, <inline-formula> <tex-math notation="LaTeX">L_{tgp}^{\alpha _{1}, \alpha _{2}} </tex-math></inline-formula> is a non-convex function, ATGP-SVM is difficult to solve. Therefore, we formulated the ATGP-SVM utilizing DC (difference of convex functions) programming and subsequently resolved it through the DC algorithm (DCA). The experimental results obtained from diverse benchmark datasets underscore the effectiveness of our proposed formulation when compared to other state-of-the-art classification models.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2024.3485214