Signal-background discrimination with convolutional neural networks in the PandaX-III experiment using MC simulation

TL;DR

This paper demonstrates that convolutional neural networks applied to 2D projections of detector tracks can significantly improve background suppression and signal efficiency in the PandaX-III neutrinoless double beta decay experiment.

Contribution

It introduces a CNN-based method for signal-background discrimination using Monte Carlo simulations, achieving over 100-fold background suppression and enhanced efficiency.

Findings

Background suppressed by a factor >100

62% improvement in efficiency ratio $\\epsilon_{s}/\sqrt{\epsilon_{b}}$

High signal efficiency maintained

Abstract

The PandaX-III experiment will search for neutrinoless double beta decay of $^{136}$Xe with high pressure gaseous time projection chambers at the China Jin-Ping underground Laboratory. The tracking feature of gaseous detectors helps suppress the background level, resulting in the improvement of the detection sensitivity. We study a method based on the convolutional neural networks to discriminate double beta decay signals against the background from high energy gammas generated by $^{214}$Bi and $^{208}$Tl decays based on detailed Monte Carlo simulation. Using the 2-dimensional projections of recorded tracks on two planes, the method successfully suppresses the background level by a factor larger than 100 with a high signal efficiency. An improvement of $62\%$ on the efficiency ratio of $\epsilon_{s}/\sqrt{\epsilon_{b}}$ is achieved in comparison with the baseline in the PandaX-III conceptual design report.