Deep-learning-based low-energy trigger algorithms for the Hyper-Kamiokande experiment
This work addresses the problem of efficiently identifying low-energy neutrino events in real-time for the Hyper-Kamiokande experiment, representing a strong specific gain over existing methods.
This paper developed deep-learning-based trigger algorithms for the Hyper-Kamiokande experiment to detect low-energy neutrino events. A supervised model achieved a signal identification efficiency of 76.7% for 3 MeV single electrons, significantly outperforming the traditional hit-count-based trigger's 26.4% efficiency.
Modern machine learning techniques have become increasingly important in particle physics because of their powerful pattern-recognition capabilities, including in real-time data acquisition where stringent runtime constraints apply. This paper details the performance of deep-learning-based trigger algorithms for a large water Cherenkov detector such as Hyper-Kamiokande aimed at low-energy neutrino events (below 7 MeV). The performance of custom neural-network supervised classifiers is shown alongside two anomaly-detection approaches trained solely on detector noise: a pure autoencoder and an energy-based model based on Manifold Projection--Diffusion Recovery (MPDR). The supervised model shows signal identification efficiencies of 76.7% for single electrons of 3 MeV kinetic energy, significantly exceeding signal efficiencies obtained from a traditional hit-count-based trigger of 26.4%, as does the MPDR approach with 31.8%. Runtime evaluations on GPU yield per-window inference latencies well below the millisecond scale, indicating that real-time operation is feasible.