The Korean Physical Society 06130 22, Teheran-ro 7-gil, Gangnam-gu, Seoul, Republic of Korea 610 Representation : Tae Won NOH TEL: 02-556-4737 FAX: 02-554-1643 E-mail : Copyright(C) KPS, All rights reserved.
30 May 2022 to 4 June 2022
Virtual Seoul
Asia/Seoul timezone

Machine Learning Methods for Solar Neutrino Classification

Not scheduled
Virtual Seoul

Virtual Seoul

Poster Solar neutrinos Poster


Alejandro Yankelevich (University of California, Irvine)


Super-Kamiokande has observed boron-8 solar neutrino recoil electrons at kinetic energies as low as 3.49 MeV to study neutrino flavor conversion within the sun. At SK-observable energies, these conversions are dominated by the Mikheyev–Smirnov–Wolfenstein effect. An upturn in the electron survival probability in which vacuum neutrino oscillations become dominant is predicted to occur at lower energies, but radioactive background increases exponentially with decreasing energy. New machine learning approaches provide substantial background reduction in the 2.49 MeV - 3.49 MeV energy region such that statistical extraction of solar neutrino interactions becomes feasible. The solar angle distributions of events selected by a ResNet convolutional neural network and a transformer network trained on event display images as well as a boosted decision tree trained on reconstructed variables used in the SK solar analysis will be presented.

Collaboration Super-Kamiokande

Primary author

Alejandro Yankelevich (University of California, Irvine)

Presentation Materials