# End-to-End Jet Classification of Quarks and Gluons with the CMS Open Data

## TL;DR

This paper develops end-to-end jet image classifiers using simulated CMS Open Data to distinguish quark-initiated from gluon-initiated jets, achieving high accuracy and robustness against pile-up effects.

## Contribution

It introduces a novel end-to-end classification approach directly from low-level detector data, outperforming traditional feature-based methods in jet discrimination tasks.

## Key findings

- Achieves competitive performance with state-of-the-art classifiers.
- Demonstrates robustness against pile-up and underlying event effects.
- Extends the approach to event-level quark vs. gluon di-jet classification.

## Abstract

We describe the construction of end-to-end jet image classifiers based on simulated low-level detector data to discriminate quark- vs. gluon-initiated jets with high-fidelity simulated CMS Open Data. We highlight the importance of precise spatial information and demonstrate competitive performance to existing state-of-the-art jet classifiers. We further generalize the end-to-end approach to event-level classification of quark vs. gluon di-jet QCD events. We compare the fully end-to-end approach to using hand-engineered features and demonstrate that the end-to-end algorithm is robust against the effects of underlying event and pile-up.