Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quantum Convolutional Neural Networks for the detection of Gamma-Ray Bursts in the AGILE space mission data

A. Rizzo, N. Parmiggiani, A. Bulgarelli, A. Macaluso, V. Fioretti, L. Castaldini, A. Di Piano, G. Panebianco, C. Pittori, M. Tavani, C. Sartori, C. Burigana, V. Cardone, F. Farsian, M. Meneghetti, G. Murante, R. Scaramella, F. Schillirò, V. Testa, T. Trombetti

TL;DR

Problem: Detect Gamma-Ray Bursts in AGILE mission data using quantum machine learning. Approach: Develop and evaluate QCNNs (fully quantum, hybrid) on sky maps and light curves, employing multiple quantum frameworks and data encodings; Data preprocessing includes resizing, smoothing, rebinning, and normalization. Findings: QCNNs achieve competitive accuracy relative to classical CNNs on sky maps with far fewer trainable parameters, while performance on light curves lags due to limited data and simulator-based training; training times are longer on simulators. Significance: Demonstrates the potential of quantum DL for high-energy astrophysics and motivates future experiments on real quantum hardware and extensions to other detectors and next-generation facilities.

Abstract

Quantum computing represents a cutting-edge frontier in artificial intelligence. It makes use of hybrid quantum-classical computation which tries to leverage quantum mechanic principles that allow us to use a different approach to deep learning classification problems. The work presented here falls within the context of the AGILE space mission, launched in 2007 by the Italian Space Agency. We implement different Quantum Convolutional Neural Networks (QCNN) that analyze data acquired by the instruments onboard AGILE to detect Gamma-Ray Bursts from sky maps or light curves. We use several frameworks such as TensorFlow-Quantum, Qiskit and PennyLane to simulate a quantum computer. We achieved an accuracy of 95.1% on sky maps with QCNNs, while the classical counterpart achieved 98.8% on the same data, using however hundreds of thousands more parameters.

Quantum Convolutional Neural Networks for the detection of Gamma-Ray Bursts in the AGILE space mission data

TL;DR

Problem: Detect Gamma-Ray Bursts in AGILE mission data using quantum machine learning. Approach: Develop and evaluate QCNNs (fully quantum, hybrid) on sky maps and light curves, employing multiple quantum frameworks and data encodings; Data preprocessing includes resizing, smoothing, rebinning, and normalization. Findings: QCNNs achieve competitive accuracy relative to classical CNNs on sky maps with far fewer trainable parameters, while performance on light curves lags due to limited data and simulator-based training; training times are longer on simulators. Significance: Demonstrates the potential of quantum DL for high-energy astrophysics and motivates future experiments on real quantum hardware and extensions to other detectors and next-generation facilities.

Abstract

Quantum computing represents a cutting-edge frontier in artificial intelligence. It makes use of hybrid quantum-classical computation which tries to leverage quantum mechanic principles that allow us to use a different approach to deep learning classification problems. The work presented here falls within the context of the AGILE space mission, launched in 2007 by the Italian Space Agency. We implement different Quantum Convolutional Neural Networks (QCNN) that analyze data acquired by the instruments onboard AGILE to detect Gamma-Ray Bursts from sky maps or light curves. We use several frameworks such as TensorFlow-Quantum, Qiskit and PennyLane to simulate a quantum computer. We achieved an accuracy of 95.1% on sky maps with QCNNs, while the classical counterpart achieved 98.8% on the same data, using however hundreds of thousands more parameters.
Paper Structure (4 sections, 2 figures, 1 table)

This paper contains 4 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methodology
  3. Results
  4. Conclusions & Future Works

Figures (2)

  • Figure 1: Example of sky maps representing background noise (on the left) and a GRB signal (on the right).
  • Figure 2: Example of a rebinned light curve representing the signal GRB100719D. The x-axis represents the time window with bins of 2.048 seconds. On the y-axis, the count rate of photons detected by the ACS over time.