Search for the 16O(ppp) → 13C π+ π+ e+ Decay Mode in Super-Kamiokande Using Machine Learning Techniques
Abstract
We report a new partial lifetime limit of 4.2 × 1032 years for the trinucleon decay mode 16O(ppp) → 13C π+ π+ e+, obtained from a search conducted using the Super-Kamiokande detector with 0.401 megaton-years of exposure across five operational periods (SK-I: 1996--2001, SK-II: 2002--2005, SK-III: 2006--2008, SK-IV: 2008--2018, SK-V: 2019--2020). This represents an improvement of six orders of magnitude over previous experimental constraints. The analysis utilizes a convolutional neural network (CNN) incorporating an attention mechanism---a computational technique that enables the model to focus on the most relevant regions of Cherenkov ring patterns---to enhance event classification, thereby improving the sensitivity of the search. This is the first application of a CNN to a nucleon decay search in Super-Kamiokande. Furthermore, the large dataset available in Super-Kamiokande (hereafter "SK") strengthens the statistical power of the study, enabling a more stringent constraint than those set by prior experiments.
Related
- Search for proton decay into a single charged antilepton and a massless invisible particle using the full pure water data set of Super-Kamiokande
- Calibration of electromagnetic shower features in the CMS calorimeter with machine-learning techniques
- Machine Learning β-decay Half-lives and Their Application to r-Process Observables
- Machine Learning Assisted Parametrisation and Prediction of Bare to Neutral β- Decay Rate Ratios of Fully Ionised Atoms
- Deep-learning-based low-energy trigger algorithms for the Hyper-Kamiokande experiment
BibTeX
@misc{feng2026search,
title = {{Search for the $^16\textO(ppp) \rightarrow ^13\textC π^+ π^+ e^+$ Decay Mode in Super-Kamiokande Using Machine Learning Techniques}},
author = {J Feng and K Abe and Y Asaoka and M Harada and Y Hayato and K Hiraide and T.H Hung and K Ieki and M Ikeda and J Kameda and Y Kataoka and S Mine and M Miura and S Moriyama and K Nakagiri and M Nakahata and S Nakayama and Y Noguchi and G Pronost and K Sato and H Sekiya and R Shinoda and M Shiozawa and Y Suzuki and A Takeda and Y Takemoto and H Tanaka and S Chen and Y Itow and T Kajita and R Nishijima and K Okumura and T Tashiro and T Tomiya and X Wang and F.J de Garay Arcones and P Fernandez and L Labarga and D Samudio and C Yanagisawa and B Jargowsky and E Kearns and J Mirabito and L Wan and T Wester and B.W Pointon and J Bian and B Cortez and N.J Griskevich and Y Jiang and M.B Smy and H.W Sobel and V Takhistov and J Hill and B.D Xu and D Jung and D.H Moon and R.G Park and B.S Yang and K Scholberg},
howpublished = {HAL},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.25681},
eprint = {2609.25681},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05780376v1},
}
RIS
TY - UNPB
TI - Search for the ^16O(ppp) → ^13C π^+ π^+ e^+ Decay Mode in Super-Kamiokande Using Machine Learning Techniques
AU - J Feng
AU - K Abe
AU - Y Asaoka
AU - M Harada
AU - Y Hayato
AU - K Hiraide
AU - T.H Hung
AU - K Ieki
AU - M Ikeda
AU - J Kameda
AU - Y Kataoka
AU - S Mine
AU - M Miura
AU - S Moriyama
AU - K Nakagiri
AU - M Nakahata
AU - S Nakayama
AU - Y Noguchi
AU - G Pronost
AU - K Sato
AU - H Sekiya
AU - R Shinoda
AU - M Shiozawa
AU - Y Suzuki
AU - A Takeda
AU - Y Takemoto
AU - H Tanaka
AU - S Chen
AU - Y Itow
AU - T Kajita
AU - R Nishijima
AU - K Okumura
AU - T Tashiro
AU - T Tomiya
AU - X Wang
AU - F.J de Garay Arcones
AU - P Fernandez
AU - L Labarga
AU - D Samudio
AU - C Yanagisawa
AU - B Jargowsky
AU - E Kearns
AU - J Mirabito
AU - L Wan
AU - T Wester
AU - B.W Pointon
AU - J Bian
AU - B Cortez
AU - N.J Griskevich
AU - Y Jiang
AU - M.B Smy
AU - H.W Sobel
AU - V Takhistov
AU - J Hill
AU - B.D Xu
AU - D Jung
AU - D.H Moon
AU - R.G Park
AU - B.S Yang
AU - K Scholberg
PY - 2026
DA - 2026/09/26
PB - HAL
DO - 10.48550/arxiv.2609.25681
UR - https://hal.science/hal-05780376v1
AB - We report a new partial lifetime limit of 4.2 × 10^32 years for the trinucleon decay mode ^16O(ppp) → ^13C π^+ π^+ e^+, obtained from a search conducted using the Super-Kamiokande detector with 0.401 megaton-years of exposure across five operational periods (SK-I: 1996--2001, SK-II: 2002--2005, SK-III: 2006--2008, SK-IV: 2008--2018, SK-V: 2019--2020). This represents an improvement of six orders of magnitude over previous experimental constraints. The analysis utilizes a convolutional neural network (CNN) incorporating an attention mechanism---a computational technique that enables the model to focus on the most relevant regions of Cherenkov ring patterns---to enhance event classification, thereby improving the sensitivity of the search. This is the first application of a CNN to a nucleon decay search in Super-Kamiokande. Furthermore, the large dataset available in Super-Kamiokande (hereafter "SK") strengthens the statistical power of the study, enabling a more stringent constraint than those set by prior experiments.
ER -