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

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.

References in corpus (34)

• Asymptotic formulae for likelihood-based tests of new physics
• Dark Matter Search Results from a One Tonne$\times$Year Exposure of XENON1T
• First results from the LUX dark matter experiment at the Sanford Underground Research Facility
• Results from a search for dark matter in the complete LUX exposure
• The frontier of simulation-based inference
• First Results from the XENON10 Dark Matter Experiment at the Gran Sasso National Laboratory
• Dark Matter Search Results from the PandaX-4T Commissioning Run
• First Direct Detection Limits on sub-GeV Dark Matter from XENON10
• Improved Limits on Scattering of Weakly Interacting Massive Particles from Reanalysis of 2013 LUX data
• The Large Underground Xenon (LUX) Experiment
• Projected WIMP sensitivity of the LUX-ZEPLIN (LZ) dark matter experiment
• Simultaneous Measurement of Ionization and Scintillation from Nuclear Recoils in Liquid Xenon as Target for a Dark Matter Experiment
• Roundtrip: A Deep Generative Neural Density Estimator
• Tritium calibration of the LUX dark matter experiment
• Lowering the radioactivity of the photomultiplier tubes for the XENON1T dark matter experiment
• Radiogenic and Muon-Induced Backgrounds in the LUX Dark Matter Detector
• The LUX-ZEPLIN (LZ) radioactivity and cleanliness control programs
• An Ultra-Low Background PMT for Liquid Xenon Detectors
• Discrimination of electronic recoils from nuclear recoils in two-phase xenon time projection chambers
• Calibration, event reconstruction, data analysis and limits calculation for the LUX dark matter experiment
• Deep Neural Networks for Energy and Position Reconstruction in EXO-200
• ${}^{83\text{m}}$Kr calibration of the 2013 LUX dark matter search
• Position Reconstruction in LUX
• Early Results on Radioactive Background Characterization for Sanford Laboratory and DUSEL Experiments
• Liquid xenon scintillation measurements and pulse shape discrimination in the LUX dark matter detector
• Modeling Pulse Characteristics in Xenon with NEST
• 3D Modeling of Electric Fields in the LUX Detector
• Scintillation-only Based Pulse Shape Discrimination for Nuclear and Electron Recoils in Liquid Xenon
• Scintillation Pulse Shape Discrimination in a Two-Phase Xenon Time Projection Chamber
• Improved Measurements of the \b{eta}-Decay Response of Liquid Xenon with the LUX Detector
• Constraints on Effective Field Theory Couplings Using 311.2 days of LUX Data
• Improved Modeling of $\beta$ Electronic Recoils in Liquid Xenon Using LUX Calibration Data
• Finding Dark Matter Faster with Explicit Profile Likelihoods
• The Design of Global Correlation Quantifiers and Continuous Notions of Statistical Sufficiency

Cited by in corpus (6)

• Constraints on cosmic-ray boosted dark matter with realistic cross section
• Detector signal characterization with a Bayesian network in XENONnT
• Insights into Dark Matter Direct Detection Experiments: Decision Trees versus Deep Learning
• Bayesian technique to combine independently-trained Machine-Learning models applied to direct dark matter detection
• FlameNEST: Explicit Profile Likelihoods with the Noble Element Simulation Technique
• Suppression of accidental backgrounds with deep neural networks in the PandaX-II experiment