Deep Learning for direct Dark Matter search with nuclear emulsions

TL;DR

This paper introduces a novel deep learning approach using 3D CNNs optimized by Bayesian search to distinguish nuclear recoil tracks from background in nuclear emulsions for dark matter detection, significantly improving background rejection.

Contribution

It presents a new deep learning method leveraging 3D CNNs and Bayesian optimization for better signal/background discrimination in dark matter searches with nuclear emulsions.

Findings

Significant improvement in background reduction over traditional methods.

Effective use of polarisation dependence of Surface Plasmon Resonance.

Enhanced signal efficiency in nuclear recoil track identification.

Abstract

We propose a new method for the discrimination of sub-micron nuclear recoil tracks from an instrumental background in fine-grain nuclear emulsions used in the directional dark matter search. The proposed method uses a 3D Convolutional Neural Network, whose parameters are optimised by Bayesian search. Unlike previous studies focused on extracting the directional information, we focus on the signal/background separation exploiting the polarisation dependence of the Localised Surface Plasmon Resonance phenomenon. Comparing the proposed method with the conventional cut-based approach shows a significant boost in the reduction factor for given signal efficiency.