Machine Learning in Gamma Astronomy

Imaging Atmospheric Cherenkov Telescopes (IACT) register extensive air showers (EAS) generated by gamma rays and cosmic rays (charged particles) when they interact with the atmosphere. These events are images recorded by IACT's highly sensitive camera, which consists of hundreds of photomultiplier tubes. Therefore, the main task of analyzing IACT data is the ability to distinguish between these two types of EAS. In addition, other properties of the primary high-energy particle, such as energy and direction of arrival, may be determined too. In recent years, deep learning methods have made significant progress in identifying gamma events and reconstructing their features. The main purpose of this paper is to provide an overview of works that use deep learning to analyze IACT data. Note that the principles of deep learning methods themselves are not considered here. They can be found, for example, in the books [1, 2] and in the reviews [3, 4, 5].