A simulation study to distinguish prompt photon from $\pi^0$ and beam halo in a granular calorimeter using deep networks

TL;DR

This study demonstrates that deep learning, specifically CNNs applied to calorimeter images, can effectively distinguish prompt photons from background sources like $ ext{π}^0$ decays and beam halo muons in collider detectors, outperforming traditional methods.

Contribution

It introduces a deep learning approach using CNNs on calorimeter images to improve prompt photon identification in collider experiments, showing significant performance gains.

Findings

Achieved 99.96% background rejection for beam halo at 99% signal efficiency.

Achieved 97.7% background rejection for $ ext{π}^0$ at 90% signal efficiency.

Deep learning outperforms traditional variable-based methods.

Abstract

In a hadron collider environment identification of prompt photons originating in a hard partonic scattering process and rejection of non-prompt photons coming from hadronic jets or from beam related sources, is the first step for study of processes with photons in final state. Photons coming from decay of $\pi_0$'s produced inside a hadronic jet and photons produced in catastrophic bremsstrahlung by beam halo muons are two major sources of non-prompt photons. In this paper the potential of deep learning methods for separating the prompt photons from beam halo and $\pi^0$'s in the electromagnetic calorimeter of a collider detector is investigated, using an approximate description of the CMS detector. It is shown that, using only calorimetric information as images with a Convolutional Neural Network, beam halo (and $\pi^{0}$) can be separated from photon with 99.96\% (97.7\%) background rejection for 99.00\% (90.0\%) signal efficiency which is much better than traditionally employed variables.