Deep-learning-based low-energy trigger algorithms for the Hyper-Kamiokande experiment

arXiv:2605.31391 · physics.ins-det, cs.LG, hep-ex · Submitted 2026-05-29 · Read on arXiv

physics.ins-det, cs.LG, hep-ex

Submitted: 2026-05-29

Updated: 2026-09-15

Comments: 18 pages, 8 figures

DOI: 10.1103/72mz-x21q

Code: https://github.com/saulam/hk-ml-trigger

License: http://creativecommons.org/licenses/by/4.0/

The gist: 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

Terminology

Abstract

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 a model based on Manifold Projection-Diffusion Recovery. 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%, while the Manifold Projection-Diffusion Recovery approach reaches 35.4% at the same operating point. Runtime evaluations on GPU yield per-window inference latencies well below the millisecond scale.

Sources

Related papers