Measurement of Atmospheric Neutrino Oscillation Parameters Using Convolutional Neural Networks with 9.3 Years of Data in IceCube DeepCore

Abstract

The DeepCore subdetector of the IceCube Neutrino Observatory provides access to neutrinos with energies above approximately 5 GeV. Data taken between 2012 and 2021 (3387 days) are utilized for an atmospheric <a:math xmlns:a=\http://www.w3.org/1998/Math/MathML\ display=\inline\>νμ</a:math> disappearance analysis that studied 150 257 neutrino-candidate events with reconstructed energies between 5 and 100 GeV. An advanced reconstruction based on a convolutional neural network is applied, providing increased signal efficiency and background suppression, resulting in a measurement with both significantly increased statistics compared to previous DeepCore oscillation results and high neutrino purity. For the normal neutrino mass ordering, the atmospheric neutrino oscillation parameters and their <c:math xmlns:c=\http://www.w3.org/1998/Math/MathML\ display=\inline\>1σ</c:math> errors are measured to be <e:math xmlns:e=\http://www.w3.org/1998/Math/MathML\ display=\inline\><e:mi mathvariant=\normal\>Δ</e:mi><e:mi mathvariant=\normal\>m</e:mi>322=2.40−0.04+0.05×10−3  eV2</e:math> and <i:math xmlns:i=\http://www.w3.org/1998/Math/MathML\ display=\inline\>sin2θ23=0.54−0.03+0.04</i:math>. The results are the most precise to date using atmospheric neutrinos, and are compatible with measurements from other neutrino detectors including long-baseline accelerator experiments.

R. Abbasi et al., Measurement of Atmospheric Neutrino Oscillation Parameters Using Convolutional Neural Networks with 9.3 Years of Data in IceCube DeepCore, Phys. Rev. Lett.