Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Pulse Shape Discrimination for Germanium Detectors using Variational Quantum Circuits

Fabrizio Napolitano

TL;DR

This paper tackles PSD in Broad Energy Germanium detectors by introducing a quantum-classical pipeline that uses a 10-qubit variational quantum circuit with amplitude encoding to process raw, 1024-sample waveforms. It demonstrates that a compact model with only $302$ trainable parameters can achieve an ROC AUC of $0.98$ and an overall accuracy of $97.1\%$, matching state-of-the-art classical pipelines that rely on large CNNs and denoising autoencoders. Key contributions include the first real-waveform QML benchmark against classical DL on BEGe data, an efficient amplitude-encoding strategy with logarithmic input compression, and a progressive circuit-growing training approach that mitigates barren plateaus. The work highlights significant practical implications for quantum-efficient detector readout and points toward a future where quantum-native sensors enable low-latency, low-power, high-fidelity event discrimination in nuclear and particle physics.

Abstract

Pulse shape discrimination (PSD) is a critical component in background rejection for neutrinoless double-beta decay and dark matter searches using Broad Energy Germanium (BEGe) detectors. To date, advanced discrimination has relied on Deep Learning approaches employing e.g. Denoising Autoencoders (DAE) and Convolutional Neural Networks (CNN). While effective, these models require tens of thousands of parameters and heavy pre-processing. In this work, we present, to the best of our knowledge, the first application of Quantum Machine Learning (QML) to real, experimental pulse waveforms from a germanium detector. We propose a quantum-classical hybrid approach using Variational Quantum Circuits (VQC) with amplitude encoding. By mapping the 1024-sample waveforms directly into a 10-qubit Hilbert space, we demonstrate that a VQC with only 302 trainable parameters achieves a receiver operating characteristic (ROC) area under the curve (AUC) of 0.98 and a global accuracy of 97.1%. This result demonstrates that even in the current Noisy Intermediate-Scale Quantum (NISQ) era, quantum models can match the performance of state-of-the-art classical baselines while reducing model complexity by over two orders of magnitude. Furthermore, we envision a scenario where future quantum sensors transmit quantum states directly to such processing units, exploiting the exponentially large Hilbert space in a natively quantum pipeline.

Pulse Shape Discrimination for Germanium Detectors using Variational Quantum Circuits

TL;DR

This paper tackles PSD in Broad Energy Germanium detectors by introducing a quantum-classical pipeline that uses a 10-qubit variational quantum circuit with amplitude encoding to process raw, 1024-sample waveforms. It demonstrates that a compact model with only trainable parameters can achieve an ROC AUC of and an overall accuracy of , matching state-of-the-art classical pipelines that rely on large CNNs and denoising autoencoders. Key contributions include the first real-waveform QML benchmark against classical DL on BEGe data, an efficient amplitude-encoding strategy with logarithmic input compression, and a progressive circuit-growing training approach that mitigates barren plateaus. The work highlights significant practical implications for quantum-efficient detector readout and points toward a future where quantum-native sensors enable low-latency, low-power, high-fidelity event discrimination in nuclear and particle physics.

Abstract

Pulse shape discrimination (PSD) is a critical component in background rejection for neutrinoless double-beta decay and dark matter searches using Broad Energy Germanium (BEGe) detectors. To date, advanced discrimination has relied on Deep Learning approaches employing e.g. Denoising Autoencoders (DAE) and Convolutional Neural Networks (CNN). While effective, these models require tens of thousands of parameters and heavy pre-processing. In this work, we present, to the best of our knowledge, the first application of Quantum Machine Learning (QML) to real, experimental pulse waveforms from a germanium detector. We propose a quantum-classical hybrid approach using Variational Quantum Circuits (VQC) with amplitude encoding. By mapping the 1024-sample waveforms directly into a 10-qubit Hilbert space, we demonstrate that a VQC with only 302 trainable parameters achieves a receiver operating characteristic (ROC) area under the curve (AUC) of 0.98 and a global accuracy of 97.1%. This result demonstrates that even in the current Noisy Intermediate-Scale Quantum (NISQ) era, quantum models can match the performance of state-of-the-art classical baselines while reducing model complexity by over two orders of magnitude. Furthermore, we envision a scenario where future quantum sensors transmit quantum states directly to such processing units, exploiting the exponentially large Hilbert space in a natively quantum pipeline.

Paper Structure

This paper contains 20 sections, 4 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methodology
  3. Dataset and Pre-processing
  4. The Classical Benchmark Pipeline
  5. Quantum Pipeline: Pre-processing and Encoding
  6. Amplitude Encoding
  7. Variational Ansatz and Readout
  8. Training Strategy
  9. Implementation and Computational Resources
  10. Results
  11. Classification Fidelity and Physics Performance
  12. Analysis of Progressive Training Dynamics
  13. Benchmarking: The Efficiency Advantage
  14. Stability and Generalization
  15. Discussion
  16. ...and 5 more sections

Figures (4)

  • Figure 1: View of the Variational Quantum Circuit employed in this work.
  • Figure 2: Validation accuracy across progressive training stages.
  • Figure 3: Loss across progressive training stages.
  • Figure 4: ROC curve illustrating the trade-off between true positive rate and false positive rate across different thresholds. The AUC of 0.98 indicates strong classification performance.