HAWC Performance Enhanced by Machine Learning in Gamma-Hadron Separation

TL;DR

This paper demonstrates that applying a Multilayer Perceptron machine learning model to HAWC data significantly improves gamma-hadron separation, increasing sensitivity and detection significance for astrophysical sources.

Contribution

The study introduces a machine learning approach, specifically a Multilayer Perceptron, that outperforms traditional methods in gamma-hadron separation for HAWC data.

Findings

19% increase in significance for Crab Nebula detection

ML approach surpasses traditional and other ML methods

Improves detector sensitivity with real and simulated data

Abstract

Improving gamma-hadron separation is one of the most effective ways to enhance the performance of ground-based gamma-ray observatories. With over a decade of continuous operation, the High-Altitude Water Cherenkov (HAWC) Observatory has contributed significantly to high-energy astrophysics. To further leverage its rich dataset, we introduce a machine learning approach for gamma-hadron separation. A Multilayer Perceptron shows the best performance, surpassing traditional and other Machine Learning based methods. This approach shows a notable improvement in the detector's sensitivity, supported by results from both simulated and real HAWC data. In particular, it achieves a 19\% increase in significance for the Crab Nebula, commonly used as a benchmark. These improvements highlight the potential of machine learning to significantly enhance the performance of HAWC and provide a valuable reference for ground-based observatories, such as Large High Altitude Air Shower Observatory (LHAASO) and the upcoming Southern Wide-field Gamma-ray Observatory (SWGO).