Using deep learning methods for IACT data analysis in gamma-ray astronomy: A review

Imaging Atmospheric Cherenkov Telescope (IACT) capture images of extensive air shower (EAS) generated by gamma rays and cosmic rays (charged particles) as they interact with the atmosphere. The much more frequent charged particle events form the main background in the search for gamma-ray sources, and therefore the success of IACT depends to a large extent on the ability to distinguish between these two types of events. A conclusion about the properties of a primary high-energy particle can be drawn from the images of the EAS which it initiated, registered by the telescope camera. In addition to the classification (gamma ray or charged particle), other properties such as energy and direction of arrival (DOA) need to be evaluated. Such EAS images can be analyzed using both traditional methods and deep learning methods based on artificial neural networks. This review aims to summarize the most common deep learning methods that are used to analyze the data collected with the help of the IACTs, including results obtained by the authors of this review, as well as provide references to original works for further in-depth study.