Strategies for Machine Learning Applied to Noisy HEP Datasets - Modular Solid State Detectors from SuperCDMS

Aug 5, 2026

Upload by Liam

Background reduction in the SuperCDMS dark matter experiment relies on identifying where each incident particle interaction occurs inside individual cryogenic solid-state detectors. Position reconstruction is performed by combining pulse-shape information across multiple phonon channels, a task suited for machine learning. Using Am-241 scan data from a SuperCDMS SNOLAB detector, statistical approaches including linear regression, artificial neural networks, and symbolic regression are compared, with simpler models showing stronger generalization on noisy, minimal data. Results suggest that architecture and training choices can mitigate overfitting, and the study is planned to repeat on a larger dataset to assess data-quality effects.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages27

Reading Time1 h 8 min

AccessFree to download

Introduction

The Cryogenic Dark Matter Search

Background reduction and fiducial cuts

Position reconstruction from phonon pulse shapes

arXiv :2404 . 10971v1 [hep-ex] 17 Apr 2024 Strategies for Machine Learning Applied to Noisy HEP Datasets: Modular Solid State Detectors from SuperCDMS P. B. Cushman1 , M. C. Fritts1 , A. D. Chambers2 , A. Roy3 , and T. Li4 1 School of Physics and Astronomy, University of Minnesota, Minneapolis, MN 55455, USA 2 Department of Physics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA 3National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA 4 Department of Computer Science and Engineering, University of Minnesota, Minneapolis, MN 55455, USA Abstract Background reduction in the SuperCDMS dark matter experiment depends on removing surface events within individual detectors by identifying the location of each incident particle interaction. Position reconstruction is achieved by combining pulse shape information over multiple phonon channels, a task well-suited to machine learning techniques. Data from an Am-241 scan of a SuperCDMS SNOLAB detector was used to study a selection of statistical approaches, including linear regression, artificial neural networks, and symbolic regression. Our results showed that simpler linear regression models were better able than artificial neural networks to generalize on such a noisy and minimal data set, but there are indications that certain architectures and training configurations can counter overfitting tendencies. This study will be repeated on a more complete SuperCDMS data set (in progress) to explore the interplay between data quality and the application of neural networks. 1 Introduction The SuperCDMS experiment [1] is a direct dark matter search performed with modular cryogenic solid-state detectors. Each detector is an ultrapure disk of germanium or silicon, roughly the size of a hockey puck. The detectors are stacked into towers of 6 each and the towers are are operated at cryogenic temperatures within a shielded cryostat inan underground lab. Interactions with incident particles releases ionization charges andathermal phonons, which can be collected on the top and bottom faces of each detector by thousands of sensors organized into multiple channels. Signal partition among the channels and the individual pulse shapes themselves provide information on the location of the interaction. The internal physics of the phonon and charge transport within the crystal is poorly understood, making it difficult to model the resulting pulse shapes and shared channel behavior from first principles. Therefore, it is useful to explore machine learning (ML) techniques, which can search broadly over a complex parameter space to reveal correlations and identify the most salient features. We report on the first study using ML to create position reconstruction algorithms in a prototype SuperCDMS detector illuminated by ain-situ movable radioactive source. This project is part of the FAIR4HEP initiative 1 which uses high-energy physics as a science driver for the development of community-wide FAIR (Findable, Accessible, Interoperable, and Reusable) frameworks [2] to advance our understanding of AI and provide new insights to apply AI techniques. Thus, while this paper presents our own exploration of the efficacy of AI as applied to a small and inherently noisy data set, the data is accessible [3], the results are preserved in a reproducible fashion in line with recent adaptation of FAIR principles for AI applications [4, 5] and researchers are encouraged to expand on our study2. A second data set with improved area coverage and higher statistics is in preparation and, in comparison with the current data set, will provide important insights into the limitations and strategies required when ML techniques are applied to increasingly more complex data sets. 2 The Cryogenic Dark Matter Search The nature of dark matter is still an outstanding mystery in particle astrophysics. Its existence is needed to explain th

FAQs

What problem does the paper address in the SuperCDMS experiment?

It addresses how to reduce background by identifying the position of particle interactions inside individual modular detectors, enabling rejection of events near detector surfaces.

How is position reconstruction performed from detector signals?

Position reconstruction combines pulse-shape information across multiple phonon channels, leveraging how signal partition and pulse morphology encode interaction location.

Which machine learning methods were evaluated on the noisy dataset?

The study evaluates statistical approaches including linear regression, artificial neural networks, and symbolic regression using Am-241 scan data from a SuperCDMS SNOLAB detector.

Aug 5, 2026

Upload by Liam