Binary Classification Based on Potentials

TL;DR

This paper presents a simple, computationally efficient binary classification method using potential functions, which can perform as well as or better than traditional SVMs.

Contribution

The paper introduces a novel, straightforward potential-based classification method that rivals or surpasses standard SVM performance.

Findings

Method matches or exceeds SVM accuracy.

Computational simplicity of the approach.

Effective for binary classification tasks.

Abstract

We introduce a simple and computationally trivial method for binary classification based on the evaluation of potential functions. We demonstrate that despite the conceptual and computational simplicity of the method its performance can match or exceed that of standard Support Vector Machine methods.