# Weakly Supervised Classification in High Energy Physics

## TL;DR

This paper introduces a weakly supervised classification method for high energy physics that relies solely on class proportions, matching fully supervised performance and reducing dependence on simulations.

## Contribution

It presents a novel weakly supervised approach that is robust to simulation mis-modeling and applicable to various learning problems in physics.

## Key findings

- Matches fully supervised classification performance
- Insensitive to simulation mis-modeling
- Applicable to diverse learning tasks

## Abstract

As machine learning algorithms become increasingly sophisticated to exploit subtle features of the data, they often become more dependent on simulations. This paper presents a new approach called weakly supervised classification in which class proportions are the only input into the machine learning algorithm. Using one of the most challenging binary classification tasks in high energy physics - quark versus gluon tagging - we show that weakly supervised classification can match the performance of fully supervised algorithms. Furthermore, by design, the new algorithm is insensitive to any mis-modeling of discriminating features in the data by the simulation. Weakly supervised classification is a general procedure that can be applied to a wide variety of learning problems to boost performance and robustness when detailed simulations are not reliable or not available.