Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2304.05830 · Physical Review Applied

Quantum-Enhanced Pattern Recognition

Giuseppe Ortolano, Carmine Napoli, Cillian Harney, Stefano Pirandola, Giuseppe Leonetti, Pauline Boucher, Elena Losero, Marco Genovese, Ivano Ruo Berchera

Published Aug 29, 2023• 28 citations
arXivHTMLPDF

Abstract

This work demonstrates experimentally that the advantage gained in sensing using quantum photonic resources can be sustained, and even amplified, through complex classical post-processing aimed at extracting relevant features. Despite the very different architectures of the classical algorithms tested here, the quantum advantage in classification performance appears to be robust and qualitatively very consistent. Thus the results argue for widespread use of quantum sensing technologies, in any field that deals with pattern recognition in large datasets.

DOI
10.1103/physrevapplied.20.024072

Related papers

Quantum Machine Learning for Defense Signal and Pattern Recognition: Proven Advantage and Practical Limits

Zuhair Ahmed

•Jan 1, 2026•0 citations

Applications in Quantum Hypothesis Testing

Giuseppe Ortolano

This Thesis work analyzes different protocols in the context of photonic quantum sensing.Quantum sensing in the optical and photonic domain is a developed field of quantum technologies that aims at improving the measurements beyond the performance of conventional techniques by exploiting quantum states of light.Our work finds its theoretical basis in Quantum Hypothesis Testing (QHT), a field that studies the discrimination among a finite set of hypotheses codified on a quantum state, by performing a measurement on it.While a rich field from a theoretical point of view, QHT counts few experimental realizations, with some notable exceptions, such as the ones related to Quantum Illumination.With the aim to partially fill this gap we present, as original results, a more applicative approach and experimental proof-of-principles to some QHT protocols, that could open interesting perspectives for real applications.The first result is the experimental realization of the Quantum reading (QR) protocol.The original QR proposal showed how quantum-correlated optical states of light can enhance the readout of classical digital data, stored in optical memories (an example are DVDs), when compared to classical sensing benchmarks.We have also demonstrated for the first time a quantum advantage in a QHT protocol applied to the monitoring of production processes, namely the identification of a deviation of the distribution of the end-products from a reference, a protocol that we labeled as Quantum Conformance Test (QCT).We finally discuss the more complex problem of pattern recognition.Pattern recognition is the task aimed at sorting images in predetermined classes, an example being the recognition of handwritten digits.The classification is done by classical processing, ranging from simple algorithms to sophisticated machine learning techniques.Regardless of the classification method, the pattern recognition performance is heavily influenced by the reliability of the images to be classified.In our analysis we demonstrate how, using the quantum sensing techniques discussed in the thesis, one can achieve a notable advantage in the recognition task, sometimes with a great amplification v of the pure sensing advantage.We validate our results experimentally showing the scalability of the quantum enhancement with the complexity of the application, not only in theory, but in practical conditions.We show how these results can be achieved using photonic correlated states, that can be routinely produced in laboratories and photon counting (PC) measurements, performed by commercial detectors.All of the protocols discussed show good resistance to experimental imperfections, notably to optical losses, a limiting factor for most of the quantum schemes.This, in conjunction with the relative simplicity of the experimental approach adopted in the realizations, strongly suggests that the protocols discussed are very promising avenues for near-term technological applications. C.2 Approximating a distribution with different initial dataset. Row Ashows the initial dataset and the best approximation for different numbers of experimental data taken, N = 150 and N = 500.Row B reports the same situation when the initial dataset is taken in a smaller interval.Row C considers a "faulty" initial dataset, having few measurements around the peak of the target distribution.The image is taken from the supplementary materials of Ref.[2]. . . . . .

•May 30, 2023•0 citations

Pattern Classification Using a Quantum System

Dan A. Ventura

We consider and compare three approaches to quantum pattern classification, presenting empirical results from simulations.

Joint International Conference on Information Sciences•Jan 1, 2002•5 citations

Quantum Convolutional Neural Networks for Visual Pattern Recognition

Niroji Thayalan, Vinayagamoorthy Balalojanan, Vithiyasahar Vigneswaran +4 more

Quantum machine learning is emerging as a promising frontier at the intersection of quantum computing and Artificial Intelligence, offering potential gains in speed, efficiency, and robustness. In this paper, a hybrid Quantum Convolutional Neural Network (QCNN) architecture is presented for visual pattern recognition tasks, leveraging quantum superposition and entanglement for enhanced feature extraction. Unlike previous studies that focus on small-scale or simplified datasets, this study systematically evaluates QCNNs across multiple benchmark datasets and provides a direct comparison with classical CNNs, highlighting practical quantum advantages such as parameter efficiency and noise-resilience. The proposed approach encodes classical image data from benchmark datasets, including$70 ~\mathrm{K} 28 \times 28$grayscale images from MNIST and FashionMNIST, and$\mathbf{6 0 K}$downsampled grayscale CIFAR-10 images, into quantum states using amplitude encoding, followed by hierarchical quantum convolution and pooling layers implemented via parametrised quantum circuits (PQCs). The model culminates in a classical dense layer for final classification. Experimental results demonstrate that the QCNN achieves competitive performance on low-dimensional grayscale datasets with 93.2 % accuracy on MNIST and 88.5% on Fashion MNIST, though it underperforms on more complex datasets like CIFAR-10 with 67.8 % accuracy due to current quantum hardware limitations. While not yet surpassing classical CNNs in raw accuracy, the proposed QCNN exhibits notable parameter efficiency, noise resilience, and theoretical potential for energy-efficient computation, highlighting its promise for future applications as quantum technologies mature.

International Conference on Advanced Research in Computing (ICARC)•Feb 18, 2026•1 citation

Enhanced hybrid quantum classical neural networks with novel encoding techniques for image recognition

Kandula Aditya Sri Krishna Raghav Chowdary, T. Janani

Applied Soft Computing•Feb 16, 2026•4 citations

Quantum-Enhanced Barcode Decoding and Pattern Recognition

Leonardo Banchi, Quntao Zhuang, Stefano Pirandola

Quantum hypothesis testing is one of the most fundamental problems in quantum information theory, with crucial implications in areas like quantum sensing, where it has been used to prove quantum advantage in a series of binary photonic protocols, e.g., for target detection or memory cell readout. In this work, we generalize this theoretical model to the multipartite setting of barcode decoding and pattern recognition. We start by defining a digital image as an array or grid of pixels, each pixel corresponding to an ensemble of quantum channels. Specializing each pixel to a black and white alphabet, we naturally define an optical model of a barcode. In this scenario, we show that the use of quantum entangled sources, combined with suitable measurements and data processing, greatly outperforms classical coherent-state strategies for the tasks of barcode data decoding and classification of black and white patterns. Moreover, introducing relevant bounds, we show that the problem of pattern recognition is significantly simpler than barcode decoding, as long as the minimum Hamming distance between images from different classes is large enough. Finally, we theoretically demonstrate the advantage of using quantum sensors for pattern recognition with the nearest-neighbor classifier, a supervised learning algorithm, and numerically verify this prediction for handwritten digit classification.

arXiv:2010.03594•Physical Review Applied•Dec 8, 2020•31 citations
This work demonstrates experimentally that the advantage gained in sensing using quantum photonic resources can be sustained, and even amplified, through complex classical post-processing aimed at extracting relevant features. Despite the very different architectures of the classical algorithms tested here, the quantum advantage in classification performance appears to be robust and qualitatively very consistent. Thus the results argue for widespread use of quantum sensing technologies, in any field that deals with pattern recognition in large datasets.
This Thesis work analyzes different protocols in the context of photonic quantum sensing.Quantum sensing in the optical and photonic domain is a developed field of quantum technologies that aims at improving the measurements beyond the performance of conventional techniques by exploiting quantum states of light.Our work finds its theoretical basis in Quantum Hypothesis Testing (QHT), a field that studies the discrimination among a finite set of hypotheses codified on a quantum state, by performing a measurement on it.While a rich field from a theoretical point of view, QHT counts few experimental realizations, with some notable exceptions, such as the ones related to Quantum Illumination.With the aim to partially fill this gap we present, as original results, a more applicative approach and experimental proof-of-principles to some QHT protocols, that could open interesting perspectives for real applications.The first result is the experimental realization of the Quantum reading (QR) protocol.The original QR proposal showed how quantum-correlated optical states of light can enhance the readout of classical digital data, stored in optical memories (an example are DVDs), when compared to classical sensing benchmarks.We have also demonstrated for the first time a quantum advantage in a QHT protocol applied to the monitoring of production processes, namely the identification of a deviation of the distribution of the end-products from a reference, a protocol that we labeled as Quantum Conformance Test (QCT).We finally discuss the more complex problem of pattern recognition.Pattern recognition is the task aimed at sorting images in predetermined classes, an example being the recognition of handwritten digits.The classification is done by classical processing, ranging from simple algorithms to sophisticated machine learning techniques.Regardless of the classification method, the pattern recognition performance is heavily influenced by the reliability of the images to be classified.In our analysis we demonstrate how, using the quantum sensing techniques discussed in the thesis, one can achieve a notable advantage in the recognition task, sometimes with a great amplification v of the pure sensing advantage.We validate our results experimentally showing the scalability of the quantum enhancement with the complexity of the application, not only in theory, but in practical conditions.We show how these results can be achieved using photonic correlated states, that can be routinely produced in laboratories and photon counting (PC) measurements, performed by commercial detectors.All of the protocols discussed show good resistance to experimental imperfections, notably to optical losses, a limiting factor for most of the quantum schemes.This, in conjunction with the relative simplicity of the experimental approach adopted in the realizations, strongly suggests that the protocols discussed are very promising avenues for near-term technological applications. C.2 Approximating a distribution with different initial dataset. Row Ashows the initial dataset and the best approximation for different numbers of experimental data taken, N = 150 and N = 500.Row B reports the same situation when the initial dataset is taken in a smaller interval.Row C considers a "faulty" initial dataset, having few measurements around the peak of the target distribution.The image is taken from the supplementary materials of Ref.[2]. . . . . .
Quantum machine learning is emerging as a promising frontier at the intersection of quantum computing and Artificial Intelligence, offering potential gains in speed, efficiency, and robustness. In this paper, a hybrid Quantum Convolutional Neural Network (QCNN) architecture is presented for visual pattern recognition tasks, leveraging quantum superposition and entanglement for enhanced feature extraction. Unlike previous studies that focus on small-scale or simplified datasets, this study systematically evaluates QCNNs across multiple benchmark datasets and provides a direct comparison with classical CNNs, highlighting practical quantum advantages such as parameter efficiency and noise-resilience. The proposed approach encodes classical image data from benchmark datasets, including70 K28×2870 ~\mathrm{K} 28 \times 2870 K28×28grayscale images from MNIST and FashionMNIST, and60K\mathbf{6 0 K}60Kdownsampled grayscale CIFAR-10 images, into quantum states using amplitude encoding, followed by hierarchical quantum convolution and pooling layers implemented via parametrised quantum circuits (PQCs). The model culminates in a classical dense layer for final classification. Experimental results demonstrate that the QCNN achieves competitive performance on low-dimensional grayscale datasets with 93.2 % accuracy on MNIST and 88.5% on Fashion MNIST, though it underperforms on more complex datasets like CIFAR-10 with 67.8 % accuracy due to current quantum hardware limitations. While not yet surpassing classical CNNs in raw accuracy, the proposed QCNN exhibits notable parameter efficiency, noise resilience, and theoretical potential for energy-efficient computation, highlighting its promise for future applications as quantum technologies mature.
Quantum hypothesis testing is one of the most fundamental problems in quantum information theory, with crucial implications in areas like quantum sensing, where it has been used to prove quantum advantage in a series of binary photonic protocols, e.g., for target detection or memory cell readout. In this work, we generalize this theoretical model to the multipartite setting of barcode decoding and pattern recognition. We start by defining a digital image as an array or grid of pixels, each pixel corresponding to an ensemble of quantum channels. Specializing each pixel to a black and white alphabet, we naturally define an optical model of a barcode. In this scenario, we show that the use of quantum entangled sources, combined with suitable measurements and data processing, greatly outperforms classical coherent-state strategies for the tasks of barcode data decoding and classification of black and white patterns. Moreover, introducing relevant bounds, we show that the problem of pattern recognition is significantly simpler than barcode decoding, as long as the minimum Hamming distance between images from different classes is large enough. Finally, we theoretically demonstrate the advantage of using quantum sensors for pattern recognition with the nearest-neighbor classifier, a supervised learning algorithm, and numerically verify this prediction for handwritten digit classification.