Xenon Signal Denoising via Supervised, Semi-Supervised, and Unsupervised Models

TL;DR

This paper introduces machine learning-based denoising models—supervised, semi-supervised, and unsupervised—to enhance energy resolution in liquid xenon detectors for neutrinoless double beta decay, outperforming traditional methods.

Contribution

It demonstrates that machine learning denoising models can significantly improve energy resolution in xenon detectors, even with limited prior signal knowledge.

Findings

Supervised model achieves <1% energy resolution.

Semi-supervised models reach ~1% resolution.

Unsupervised model attains ~1.5% resolution.

Abstract

This study presents a denoising algorithm trained using machine learning to improve the energy resolution of a single-phase liquid xenon time projection chamber for neutrinoless double beta decay detection. Supervised, unsupervised, and semi-supervised models are demonstrated to significantly remove noise from simulated measurements while preserving signal information. The supervised model achieves an energy resolution of $<1\%$, while the semi-supervised models achieve energy resolutions of $\sim 1\%$, and the unsupervised model performance is $\sim 1.5\%$. This work is evidence that machine learning denoising can improve energy resolution compared to traditional algorithms, even when experimentalists lack perfect a priori knowledge of the signals. Such models provide a realistic path toward next-generation sensitivity in $0\nu\beta\beta$ searches.