Fast and Flexible Analysis of Direct Dark Matter Search Data with Machine Learning
Authors:
LUX Collaboration,
D. S. Akerib,
S. Alsum,
H. M. Araújo,
X. Bai,
J. Balajthy,
J. Bang,
A. Baxter,
E. P. Bernard,
A. Bernstein,
T. P. Biesiadzinski,
E. M. Boulton,
B. Boxer,
P. Brás,
S. Burdin,
D. Byram,
N. Carrara,
M. C. Carmona-Benitez,
C. Chan,
J. E. Cutter,
L. de Viveiros,
E. Druszkiewicz,
J. Ernst,
A. Fan,
S. Fiorucci
, et al. (75 additional authors not shown)
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 betwe…
▽ More
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.
△ Less
Submitted 17 October, 2022; v1 submitted 14 January, 2022;
originally announced January 2022.
