Linear classifier, least-squares cost function, and outliers

TL;DR

This paper discusses how outliers affect linear classifiers with least-squares cost functions and proposes a simple scaling method to mitigate their negative impact, supported by numerical results.

Contribution

It introduces a straightforward scaling technique to reduce outlier influence on linear classifiers using least-squares, improving decision boundary robustness.

Findings

Outliers negatively impact the decision boundary of linear classifiers.

Simple scaling can lessen the influence of outliers.

Numerical results demonstrate improved classification boundaries with scaling.

Abstract

A set of introductory notes on the subject of data classification using a linear classifier and least-squares cost function, and the negative effect of the presence of outliers on the decision boundary of the linear discriminant. We also show how a simple scaling could make the outlier less significant, thereby obtaining a much better decision boundary. We present some numerical results.