A semi-supervised approach to dark matter searches in direct detection data with machine learning

Abstract

The dark matter sector remains completely unknown. It is therefore crucial to keep an open mind regarding its nature and possible interactions. Focusing on the case of Weakly Interacting Massive Particles, in this work we make this general philosophy more concrete by applying modern machine learning techniques to dark matter direct detection. We do this by encoding and decoding the graphical representation of background events in the XENONnT experiment with a convolutional variational autoencoder. We describe a methodology that utilizes the `anomaly score' derived from the reconstruction loss of the convolutional variational autoencoder as well as a pre-trained standard convolutional neural network, in a semi-supervised fashion. Indeed, we observe that optimum results are obtained only when both unsupervised and supervised anomaly scores are considered together. A data set that has a higher proportion of anomaly score is deemed anomalous and deserves further investigation. Contrary to classical analyses, in principle all information about the events is used, preventing unnecessary information loss. Lastly, we demonstrate the reach of learning-focused anomaly detection in this context by comparing results with classical inference, observing that, if tuned properly, these techniques have the potential to outperform likelihood-based methods.

Matches published version in JCAP

References in corpus (10)

• Deep Learning in Neural Networks: An Overview
• Dark Matter Results From 54-Ton-Day Exposure of PandaX-II Experiment
• DARWIN: towards the ultimate dark matter detector
• Extending the Bump Hunt with Machine Learning
• First results on dark matter annual modulation from ANAIS-112 experiment
• Conceptual design and simulation of a water Cherenkov muon veto for the XENON1T experiment
• Challenges for Unsupervised Anomaly Detection in Particle Physics
• The ArDM Liquid Argon Time Projection Chamber at the Canfranc Underground Laboratory: a ton-scale detector for Dark Matter Searches
• From Dark Matter to Galaxies with Convolutional Networks
• Improved quality tests of R11410-21 photomultiplier tubes for the XENONnT experiment

Cited by in corpus (7)

• Deep Generative Models for Detector Signature Simulation: A Taxonomic Review
• Anomaly Detection under Coordinate Transformations
• Improving ANAIS-112 sensitivity to DAMA/LIBRA signal with machine learning techniques
• Non-resonant Anomaly Detection with Background Extrapolation
• 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
• Assessing Galaxy Rotation Kinematics: Insights from Convolutional Neural Networks on Velocity Variations