Application of Machine Learning Methods in the Baikal-GVD Experiment
https://doi.org/10.56304/S2079562923010116
Abstract
The Baikal-GVD experiment is a neutrino telescope located in Lake Baikal, Russia. As of 2022, it has an effective volume of 0.5 km3, which makes it the largest neutrino telescope in the Northern Hemisphere and the second largest in the world. This article presents an overview of machine learning methods developed to analyze data from the Baikal-GVD experiment. Specifically, we discuss neural networks developed to (1) suppress noise responses of optical modules, (2) identify neutrino-induced events and estimate their flux, and (3) recover the neutrino arrival angle. It is shown that neural networks are comparable or superior in accuracy to standard algorithmic event reconstruction procedures for similar problems.
Supporting agencies: This work was supported by the Russian Science Foundation, project no. 22-22-20063.
About the Authors
I. V. Kharuk
Institute for Nuclear Research, Russian Academy of Sciences, Moscow, 117312 Russia
Moscow Institute of Physics and Technology, Dolgoprudny, 141700 Russia
Russian Federation
Russian Federation
А. V. Matseiko
Institute for Nuclear Research, Russian Academy of Sciences, Moscow, 117312 Russia
Moscow Institute of Physics and Technology, Dolgoprudny, 141700 Russia
Russian Federation
Russian Federation
А. Yu. Leonov
Institute for Nuclear Research, Russian Academy of Sciences, Moscow, 117312 Russia
Moscow Institute of Physics and Technology, Dolgoprudny, 141700 Russia
Russian Federation
Russian Federation
References
1. <em>Aartsen M.G. et al.</em> // Science. 2013. V. 342. P. 1242856. https://arxiv.org/abs/1311.5238.
2. <em>Belolaptikov I. et al.</em> // Proc. PoS ICRC2021. P. 002. https://arxiv.org/abs/2109.14344.
3. <em>Allakhverdyan V.A. et al.</em> // Phys. Rev. D. 2023. V. 107. P. 042005. https://arxiv.org/abs/2211.09447.
4. <em>Aiello S. et al.</em> // Astropart. Phys. 2019. V. 111. P. 100−110. https://arxiv.org/abs/1810.08499.
5. <em>Aartsen M.G., Abbasi R., Ackermann M. et al.</em> // J. Phys. G. 2021. V. 48. P. 060501.
6. <em>Malyshkin Y. et al.</em> // Nucl. Instrum. Methods. Phys. Res. B. 2023. V. 1050. P. 168117.
7. <em>Choma N., Monti F., Gerhardt L., et al.</em> // https://arxiv.org/abs/1809.06166.
8. <em>Huennefeld M.</em> // Proc. PoS ICRC2017. P. 1057.
9. <em>Huennefeld M.</em> // EPJ Web of Conf. 2019. V. 207. P. 05005.
10. <em>Reck S., Guderian D., Vermarien G., Domi A.</em> // J. Instrum. 2021. V. 16. P. C10011. https://arxiv.org/abs/2107.13375.
11. <em>Aiello S., Albert A., Garre S.A., et al.</em> // J. Instrum. 2020. V. 15. P. P10005.
12. <em>Ronneberger O., Fischer P., Brox T.</em> // Proc. Intl. Conf. on Medical Image Computing and Computer-Assisted Intervention. P. 234−241.
13. <em>Avrorin A.D. et al.</em> // Proc. PoS ICRC2021 P. 1063. https://arxiv.org/abs/2108.00208.
14. <em>Lin T.-Y., Goyal P., Girshick R., He K., Doll’ar P.</em> // Proc. IEEE Intl. Conf. on Computer Vision. P. 2980−2988.
15. <em>Wang Y., Sun Y., Liu Z., Sarma S.E., Bronstein M.M., Solomon J.M.</em> // ACM Trans. Graph. 2019. V. 38. P. 1−12.
Review
For citations:
Kharuk I.V., Matseiko А.V., Leonov А.Yu. Application of Machine Learning Methods in the Baikal-GVD Experiment. Nuclear Physics and Engineering. 2024;15(1):36-42. (In Russ.) https://doi.org/10.56304/S2079562923010116
Views:
29
Email this article 