Abstract
Baikal-GVD is a large (\(\sim\)1 km\({}^{3}\)) underwater neutrino telescope located in Lake Baikal, Russia. In this report, we present two machine learning techniques developed for its data analysis. First, we introduce a neural network for an efficient rejection of noise hits, emerging due to natural water luminescence. Second, we develop a neural network for distinguishing muon- and neutrino-induced events. By choosing an appropriate classification threshold, we preserve \(90\%\) of neutrino-induced events, while muon-induced events are suppressed by a factor of \(10^{-6}\). Both of the developed neural networks employ the causal structure of events and surpass the precision of standard algorithmic approaches.
REFERENCES
IceCube Collab., Science 342, 1242856 (2013). https://doi.org/10.1126/science.1242856
Baikal-GVD Collab., in Proc. 37th Int. Cosmic Ray Conf., Berlin, 2021 (PoS, Trieste, 2021), Vol. 395, p. 2. https://doi.org/10.22323/1.395.0002
S. Aiello et al. (KM3NeT collab.), Astropart. Phys. 111, 100 (2019). https://doi.org/10.1016/j.astropartphys.2019.04.002
M. G. Aartsen et al. (IceCube Collab.), J. Phys. G: Nucl. Part. Phys. 48, 060501 (2021). https://doi.org/10.1088/1361-6471/abbd48
V. A. Allakhverdyan et al. (Baikal-GVD Collab.), Phys. Rev. D 107, 042005 (2023). https://doi.org/10.1103/PhysRevD.107.042005
Y. Malyshkin et al. (Baikal-GVD Collab.), Nucl. Instrum. Methods Phys. Res., Sect. A 1050, 168117 (2023). https://doi.org/10.1016/j.nima.2023.168117
N. Choma et al. (IceCube Collab.), in Proc. 17th IEEE Int. Conf. on Machine Learning and Applications, Orlando, Fla., 2018 (IEEE, 2019), pp. 386–391. https://doi.org/10.1109/ICMLA.2018.00064
M. Huennefeld et al. (IceCube Collab.), in 35th Int. Cosmic Ray Conf., Busan, Korea, 2017 (PoS, Trieste, 2018), Vol. 301, p. 1057. https://doi.org/10.22323/1.301.1057
M. Huennefeld, EPJ Web Conf. 207, 05005 (2019). https://doi.org/10.1051/epjconf/201920705005
S. Reck et al. (KM3NeT Collab.), J. Instrum. 16, C10011 (2021). https://doi.org/10.1088/1748-0221/16/10/C10011
S. Aiello et al. (KM3NeT Collab.), J. Instrum. 15, P10005 (2020). https://doi.org/10.1088/1748-0221/15/10/P10005
J. García-Méndez et al. (ANTARES Collab.), J. Instrum. 16, C09018 (2021). https://doi.org/10.1088/1748-0221/16/09/C09018
The IceCube Collab., J. Instrum. 16, P07041 (2021). https://doi.org/10.1088/1748-0221/16/07/P07041
A. D. Avrorin et al. (Baikal-GVD Collab.), in Proc. 36th Int. Cosmic Ray Conf., Madison, Wis., 2019 (PoS, Trieste, 2021), Vol. 358, p. 875. https://doi.org/10.22323/1.358.0875
N. N. Kalmykov and S. S. Ostapchenko, Phys. At. Nucl. 56, 346 (1993).
D. Heck, J. Knapp, J. N. Capdevielle, et al., CORSIKA: A Monte Carlo Code to Simulate Extensive Air Showers (Forschungszentrum Karlsruhe, Karlsruhe, Germany, 1998).
L. Dominé, et al. (DeepLearnPhysics Collab.), Phys. Rev. D 104, 032004 (2021). https://doi.org/10.1103/PhysRevD.104.032004
J. Long, E. Shelhamer, and T. Darrell, in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Boston, 2015 (IEEE, 2015), pp. 3431–3440. https://doi.org/10.1109/CVPR.2015.7298965
O. Ronneberger, P. Fischer and T. Brox, in Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015, Ed. by N. Navab, J. Hornegger, W. Wells, and A. Frangi, Lecture Notes in Computer Science, Vol. 9351 (Springer, Cham, 2015), pp. 234–241. https://doi.org/10.1007/978-3-319-24574-4_28
Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang, in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. DLMIA ML-CDS 2018, Ed. by D. Stoyanov et al., Lecture Notes in Computer Sciences, Vol. 11045 (Springer, Cham, 2018), pp. 3–11. https://doi.org/10.1007/978-3-030-00889-5_1
Pierre Auger Collab., J. Instrum. 16, P07016 (2021). https://doi.org/10.1088/1748-0221/16/07/P07016
D. Bahdanau, K. Cho, and Y. Bengio, ‘‘Neural machine translation by jointly learning to align and translate,’’ arXiv Preprint (2014). https://doi.org/10.48550/arXiv.1409.0473
I. Sutskever, O. Vinyals, and Q. V. Le, in Proc. 27th Int. Conf. on Neural Information Processing Systems, Montreal, Canada, 2014 (MIT Press, Cambridge, 2014), Vol. 2, pp. 3104–3112. https://doi.org/10.48550/arXiv.1409.3215
S. Hochreiter and J. Schmidhuber, Neural Comput. 9, 1735 (1997). https://doi.org/10.1162/neco.1997.9.8.1735
B. Shaybonov et al. (Baikal-GVD Collab.), in Proc. 37th Int. Cosmic Ray Conf., Berlin, 2021 (PoS, Trieste, 2021), Vol. 395, p. 1063. https://doi.org/10.22323/1.395.1063
K. He, X. Zhang, S. Ren and J. Sun, in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Las Vegas, 2016 (IEEE, 2016), pp. 770–778. https://doi.org/10.1109/CVPR.2016.90
T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, IEEE Trans. Pattern Anal. Mach. Intell. 42, 318–327 (2018). https://doi.org/10.1109/TPAMI.2018.2858826.
Funding
This work was supported by the Russian Science Foundation, grant no. 22-22-20063.
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors declare that they have no conflicts of interest.
Additional information
Publisher’s Note.
Allerton Press remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
APPENDIX
APPENDIX
1.1 DERIVATION OF Eq. (3)
In this section we derive distributions, expected values and dispersions of random values in Eq. (2). Then, using them, we evaluate the error of estimating the number of neutrino events using Eq. (2). In this section, we use \(P\) to denote probability of some outcome for a random variable, and \(M\) and \(D\) for its expected value and dispersion, accordingly.
We start by discussing the properties of random variables in the experimental dataset. Let the latter contains \(n^{0}\) events in total, \(n_{\nu}^{0}\) of which are neutrino-induced and \(n_{\mu}^{0}\equiv n^{0}-n_{\nu}^{0}\) are EAS-induced. \(n_{\nu}^{0}\) and \(n_{\mu}^{0}\) are random variables distributed according to the Poisson law with parameters \(\nu\) and \(\mu\), respectively. Hence,
Since \(n^{0}\) is a sum of \(n^{0}_{\nu}\) and \(n^{0}_{\mu}\), it also follows the Poisson distribution,
Its expected value and dispersion are:
Let us now address the classification of events by the neural network. A trained neural network can be considered as a black box. As it was discussed in Subsection 4.3, for a fixed classification threshold \(\xi\), the network classifies a neutrino-induced event correctly with some probability \(E\), and EAS-induced event is identified incorrectly with the probability \(S\). Hence, the number of identified true and false neutrino-induced events are independent random variables with binomial distributions:
Here \(n_{\nu}(\xi)\equiv n_{\nu}\), \(n_{\mu}(\xi)\equiv n_{\mu}\), and \(Bin(m,p)(k)\) stands for the binomial distribution with number of experiments \(m\) and success probability \(p\):
The number of neutrino-induced events identified by the neural network on a test dataset is subject to both of the above-described random processes. Hence, the full probability distributions, \(P_{\text{f}}\), of \(n_{\nu}\) and \(n_{\mu}\) can be obtained by multiplying the corresponding Poisson and binomial distributions,
Using Eqs. (A8), (A9), (A1), and Eq. (A2), one can evaluate the expected values and dispersions of \(n_{\nu}\) and \(n_{\mu}\):
For the random variable \(n\), which is a sum of \(n_{\nu}\) and \(n_{\mu}\), one has:
Next, let us elaborate on the distribution of \(\tilde{E}\) and \(\tilde{S}\) considered as random variables. Let the test Monte Carlo dataset contains \(\tilde{n}^{0}\) events, \(\tilde{n}_{\nu}^{0}\) of which are neutrino-induced and \(\tilde{n}_{\mu}^{0}\) are EAS-induced. Since the measured exposure and suppression, \(\tilde{E}\) and \(\tilde{S}\), are given by Eq. (1), they follow the probability distributions
where \(\alpha\) and \(\beta\) are discrete variables so that \(\alpha\tilde{n}_{\nu}\) and \(\beta\tilde{n}_{\mu}\) are integers. Hence,
proving that \(\tilde{E}\) and \(\tilde{S}\) are unbiased estimates of true values of \(E\) and \(S\), respectively.
Now, we are ready to evaluate the dispersion of \(N_{\xi}\) estimated using Eq. (2). For this purpose, we used the standard formula for the dispersion of a function of random variables,
Here, \(v\) and \(u\) denote arguments of the function \(N_{\xi}\), which are \(n(\xi)\equiv n\), \(n(0)\equiv n^{0}\), \(\tilde{E}(\xi)\) and \(\tilde{S}(\xi)\); \(\sigma_{v}^{2}\) stands for squared variance of \(v\), and \(Cov_{v,u}\) denotes covariance between \(v\) and \(u\).
Let us explicitly write out the estimation of the variances and covariances. According to Eq. (A4), \(\sigma_{n^{0}}^{2}\) can be estimated as
Further, from Eq. (A12), one gets
Next, \(\sigma_{\tilde{E}}^{2}\) and \(\sigma_{\tilde{S}}^{2}\) can be obtained from Eq. (A15) and Eq. (A16),
Finally, note that there are only two dependent random variables in Eq. (2)—\(n\) and \(n^{0}\). Their covariance can be calculated using Eqs. (A3), (A8), and (A9),
Therefore this covariance can be estimated as \(n\).
By calculating the partial derivatives of \(N_{\xi}\) in Eq. (A17) and substituting the obtained expressions for variances and covariances, we arrive at the final result:
About this article
Cite this article
Matseiko, A.V., Kharuk, I.V. Application of Machine Learning Methods in Baikal-GVD: Background Noise Rejection and Selection of Neutrino-Induced Events. Moscow Univ. Phys. 78 (Suppl 1), S71–S79 (2023). https://doi.org/10.3103/S0027134923070226
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S0027134923070226