# Probing Neural Networks for the Gamma/Hadron Separation of the Cherenkov Telescope Array

## TL;DR

This paper compares convolutional neural networks and boosted decision trees for gamma/hadron separation in Cherenkov Telescope Array data, finding neural networks can outperform classical methods under certain conditions.

## Contribution

It provides a comparative analysis of neural networks versus classical decision trees for particle shower classification in CTA simulations, highlighting conditions where neural networks excel.

## Key findings

- Neural networks can outperform decision trees in gamma/hadron separation.
- ROC curve analysis shows neural networks have better discrimination under specific conditions.
- The study offers insights into the effectiveness of deep learning in astrophysical data analysis.

## Abstract

We compared convolutional neural networks to the classical boosted decision trees for the separation of atmospheric particle showers generated by gamma rays from the particle-induced background. We conduct the comparison of the two techniques applied to simulated observation data from the Cherenkov Telescope Array. We then looked at the Receiver Operating Characteristics (ROC) curves produced by the two approaches and discuss the similarities and differences between both. We found that neural networks overperformed classical techniques under specific conditions.