AI-Based Models in INO-ICAL: An Overview of Applications and Studies

Measurement of neutrino mixing parameters using a magnetized iron calorimeter (ICAL) is the primary goal of INO. Until recently, most of INO-ICAL related analyses, using prototype detector data and simulations were based on conventional algorithms. However, in the last decade, AI-based analysis have shown impressive performance in many high-energy physics experiments. In this presentation, we provide an overview of machine learning (ML) algorithms that we have developed for selected analyses related to INO-ICAL and prototype detectors. The studies include directionality and charge identification, energy reconstruction of cosmic muons in mICAL, prediction of muon multiplicity, and search for new event topologies.