Fast and Flexible Analysis of Direct Dark Matter Search Data with Machine Learning

TL;DR

This paper introduces a machine learning-enhanced analysis method for dark matter search data that significantly reduces computation time, improves flexibility in variable correlation modeling, and enables the inclusion of additional variables without extra computational costs.

Contribution

The paper presents a novel machine learning approach integrated with profile likelihood fitting that accelerates analysis, captures complex correlations, and scales efficiently for future dark matter experiments.

Findings

Achieved 30-fold reduction in computation time.

Maintained performance with and without position-corrections.

Enabled inclusion of additional variables like pulse shape without increased computational burden.

Abstract

We present the results from combining machine learning with the profile likelihood fit procedure, using data from the Large Underground Xenon (LUX) dark matter experiment. This approach demonstrates reduction in computation time by a factor of 30 when compared with the previous approach, without loss of performance on real data. We establish its flexibility to capture non-linear correlations between variables (such as smearing in light and charge signals due to position variation) by achieving equal performance using pulse areas with and without position-corrections applied. Its efficiency and scalability furthermore enables searching for dark matter using additional variables without significant computational burden. We demonstrate this by including a light signal pulse shape variable alongside more traditional inputs such as light and charge signal strengths. This technique can be exploited by future dark matter experiments to make use of additional information, reduce computational resources needed for signal searches and simulations, and make inclusion of physical nuisance parameters in fits tractable.