uBoost: A boosting method for producing uniform selection efficiencies from multivariate classifiers

TL;DR

This paper introduces uBoost, a novel boosting technique that ensures uniform selection efficiency across a multivariate space, enhancing analyses where uniformity is crucial.

Contribution

uBoost is a new boosting method that achieves uniform efficiency in multivariate classifiers, tailored for specialized particle physics analyses.

Findings

Produces uniform selection efficiency across multivariate space

Suitable for amplitude analyses and similar applications

Enhances performance of neural networks and decision trees

Abstract

The use of multivariate classifiers, especially neural networks and decision trees, has become commonplace in particle physics. Typically, a series of classifiers is trained rather than just one to enhance the performance; this is known as boosting. This paper presents a novel method of boosting that produces a uniform selection efficiency in a user-defined multivariate space. Such a technique is ideally suited for amplitude analyses or other situations where optimizing a single integrated figure of merit is not what is desired.