Analyzing {\gamma}-rays of the Galactic Center with Deep Learning

TL;DR

This paper introduces a deep learning approach using convolutional neural networks to analyze Fermi-LAT gamma-ray data from the Galactic Center, aiming to distinguish between dark matter annihilation signals and unresolved point sources.

Contribution

The study develops and applies a novel deep learning method to interpret gamma-ray data, enabling precise characterization of unresolved point source contributions in the Galactic Center.

Findings

Deep learning effectively differentiates between diffuse emission and point sources.

The method constrains the contribution of unresolved sources to the gamma-ray excess.

Results support the presence of unresolved point sources as a significant component.

Abstract

We present a new method to interpret the $\gamma$-ray data of our inner Galaxy as measured by the Fermi Large Area Telescope (Fermi LAT). We train and test convolutional neural networks with simulated Fermi-LAT images based on models tuned to real data. We use this method to investigate the origin of an excess emission of GeV $\gamma$-rays seen in previous studies. Interpretations of this excess include $\gamma$ rays created by the annihilation of dark matter particles and $\gamma$ rays originating from a collection of unresolved point sources, such as millisecond pulsars. Our new method allows precise measurements of the contribution and properties of an unresolved population of $\gamma$-ray point sources in the interstellar diffuse emission model.