BSVM: A Banded Suport Vector Machine

TL;DR

The paper introduces Banded SVM (B-SVM), a new binary classification method that encourages decision rules within a specified range, resulting in two sets of support vectors near boundaries and inside classes.

Contribution

It proposes a novel Banded SVM formulation with a penalty term to control decision rule range, adding a second set of support vectors inside classes.

Findings

B-SVM effectively constrains decision rules within a user-defined interval.

The method produces two support vector sets: near boundaries and inside classes.

B-SVM improves classification control over standard SVMs.

Abstract

We describe a novel binary classification technique called Banded SVM (B-SVM). In the standard C-SVM formulation of Cortes et al. (1995), the decision rule is encouraged to lie in the interval [1, \infty]. The new B-SVM objective function contains a penalty term that encourages the decision rule to lie in a user specified range [\rho_1, \rho_2]. In addition to the standard set of support vectors (SVs) near the class boundaries, B-SVM results in a second set of SVs in the interior of each class.