Cats vs Dogs, Photons vs Hadrons

TL;DR

This paper develops a convolutional neural network for particle classification in gamma ray astronomy, demonstrating improved accuracy over traditional methods using Monte Carlo simulation data.

Contribution

It introduces a CNN approach tailored for binary particle classification in Cherenkov telescope data, outperforming classical random forest techniques.

Findings

CNN achieves higher discriminant power than random forests.

The method effectively classifies particles using uncleaned Monte Carlo images.

Results suggest potential for improved particle identification in gamma ray astronomy.

Abstract

In gamma ray astronomy with Cherenkov telescopes, machine learning models are needed to guess what kind of particles generated the detected light, and their energies and directions. The focus in this work is on the classification task, training a simple convolutional neural network suitable for binary classification (as it could be a cats vs dogs classification problem), using as input uncleaned images generated by Montecarlo data for a single ASTRI telescope. Results show an enhanced discriminant power with respect to classical random forest methods.