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:2408.11322 · Machine Learning Science and Technology

Transfer learning and the early estimation of single-photon source quality using machine learning methods

David Jacob Kedziora, Anna Musiał, Wojciech Rudno-Rudziński, Bogdan Gabryś

Published Apr 2, 2025• 2 citations
arXivHTMLPDF

Abstract

Abstract The use of single-photon sources (SPSs) is central to numerous systems and devices proposed amidst a modern surge in quantum technology. However, manufacturing schemes remain imperfect, and single-photon emission purity must often be experimentally verified via interferometry. Such a process is typically slow and costly, which has motivated growing research into whether SPS quality can be more rapidly inferred from incomplete emission statistics. Hence, this study is a sequel to previous work that demonstrated significant uncertainty in the standard method of quality estimation, i.e. the least-squares fitting of a physically motivated function, and asks: can machine learning (ML) do better? The study leverages eight datasets obtained from measurements involving an exemplary quantum emitter, i.e. a single InGaAs/GaAs epitaxial quantum dot; these eight contexts predominantly vary in the intensity of the exciting laser. Specifically, via a form of ‘transfer learning’, five ML models, three linear and two ensemble-based, are trained on data from seven of the contexts and tested on the eighth. Validation metrics quickly reveal that even a linear regressor can outperform standard fitting when it is tested on the same contexts it was trained on, but the success of transfer learning is less assured, even though statistical analysis, made possible by data augmentation, suggests its superiority as an early estimator. Accordingly, the study concludes by discussing future strategies for grappling with the problem of SPS context dissimilarity, e.g. feature engineering and model adaptation.

DOI
10.1088/2632-2153/adc86f

Related papers

Harnessing data augmentation to quantify uncertainty in the early estimation of single-photon source quality

David Jacob Kedziora, Anna Musiał, Wojciech Rudno-Rudziński +1 more

Abstract Novel methods for rapidly estimating single-photon source (SPS) quality have been promoted in recent literature to address the expensive and time-consuming nature of experimental validation via intensity interferometry. However, the frequent lack of uncertainty discussions and reproducible details raises concerns about their reliability. This study investigates the use of data augmentation, a machine learning technique, to supplement experimental data with bootstrapped samples and quantify the uncertainty of such estimates. Eight datasets obtained from measurements involving a single InGaAs/GaAs epitaxial quantum dot serve as a proof-of-principle example. Analysis of one of the SPS quality metrics derived from efficient histogram fitting of the synthetic samples, i.e. the probability of multi-photon emission events, reveals significant uncertainty contributed by stochastic variability in the Poisson processes that describe detection rates. Ignoring this source of error risks severe overconfidence in both early quality estimates and claims for state-of-the-art SPS devices. Additionally, this study finds that standard least-squares fitting is comparable to using a Poisson likelihood, and expanding averages show some promise for early estimation. Also, reducing background counts improves fitting accuracy but does not address the Poisson-process variability. Ultimately, data augmentation demonstrates its value in supplementing physical experiments; its benefit here is to emphasise the need for a cautious assessment of SPS quality.

arXiv:2306.15683•Machine Learning Science and Technology•Nov 16, 2023•3 citations

Statistical and Machine Learning Methods for Quantum Measurements with Single Photon Emitters

Dylan G. Stone

With wide applications ranging from quantum communication and metrology to biomedicine, single photon sources in solid-state hosts have become a major area of study. Here, we focus on three applications: nanothermometry, optically detected magnetic resonance (ODMR), and second order autocorrelation. We present novel statistical and machine learning (ML) approaches to extract information from experimental and simulated data and benchmark these methods against traditional inference-based statistical approaches. We found that compared to traditional inference-based methods ML algorithms can: i) predict temperatures at the nanoscale with greater accuracy and with less calibration points than traditional fitting methods; ii) identify the resonance peaks in ODMR spectra with factors ~1.3x and ~4.7x better accuracy and resolution and achieved equal or better performance with ~5x less data; and iii) have the potential to parse second order autocorrelation data more efficiently. ML algorithms are thus powerful tools for quantum sensing techniques. Author Keywords: colour centers, machine learning, nanosensing, nanothermometry, optically detected magnetic resonance, second order autocorrelation

•Jan 1, 2025•0 citations

Efficient classical training of model-free quantum photonic reservoir

Rosario Di Bartolo, Valeria Cimini, Giorgio Minati +10 more

Model-independent estimation of the properties of quantum states is a central challenge in quantum technologies, as experimental imperfections, drifts, and imprecise models of the actual quantum dynamics inevitably hinder accurate reconstructions. Here, we introduce a training strategy for photonic quantum extreme learning machines in which both the learning stage and the optimization of the measurement settings are performed entirely with classical light, while inference is carried out on genuinely quantum states. The protocol is based on the identity between the normalized output intensities following the evolution of coherent states through a linear optical reservoir, and the output statistics obtained with separable input quantum states. Building on this correspondence, we implemented a model-free, gradient-based optimization of the reservoir measurement projection directly on experimental data, without relying on a prior model of the device transformation. We experimentally show that the resulting classical-to-quantum transfer enables accurate reconstruction of single-qubit Pauli observables for previously unseen single-photon states, and extends to the estimation of a two-qubit entanglement witness for arbitrary bipartite states. Beyond demonstrating a qualitatively distinct form of out-of-distribution generalization across the classical-to-quantum boundary, our results identify a practical route to fast, adaptive, and resource-efficient training of photonic quantum learning devices.

quant-pharXiv:2604.12441•Apr 15, 2026•0 citations

Rapid photon bunching characterization via photon-number-resolved detection

Amilli Zárate-Calatayud, Samuel Corona-Aquino, Sarahí Vázquez-Hernández +5 more

Characterization of the photon statistics of an optical field is central to identifying light sources in various quantum-enabled technologies, from super-resolved quantum imaging to precise quantum sensing and spectroscopy. In practice, however, estimating photon-correlation functions with sufficient accuracy requires long acquisition times and computationally intensive statistical estimators, severely limiting real-time operation and the study of rapidly varying or weak sources. Here, we show that learned inference based on artificial neural networks enables rapid estimation of second-order coherence directly from photon-number-resolved measurements, outperforming the precision of conventional statistical methods with processing times three orders of magnitude shorter. Our approach is source-agnostic and remains accurate in few-photon illumination regimes where the statistics of coherent, partially coherent, thermal, and super-thermal light become nearly indistinguishable. We experimentally validate the method across more than 4000 distinct classical light sources, demonstrating equal or improved precision with drastically reduced data and computational requirements. These results establish machine-learning-enabled photon-statistics inference as a new operational route to rapid optical-coherence diagnostics, enabling real-time monitoring of classical light sources in regimes previously inaccessible to conventional estimators.

Applied Physics Reviews•Sep 1, 2026•0 citations

Machine Learning for Integrated Quantum Photonics

Zhaxylyk A. Kudyshev, Vladimir M. Shalaev, Alexandra E. Boltasseva

Abstract Realization of integrated quantum photonics is a key step toward scalable quantum applications such as quantum computing, sensing, information processing, and quantum material metrology. To enable practical quantum photonic systems, several challenges should be addressed, including (i) the realization of deterministic, bright, and stable single-photon emission operating at THz rates and at room temperatures, (ii) on-chip integration of efficient single-photon sources, and (iii) the development of deterministic and scalable nanoassembly of quantum circuitry elements. In this Perspective, we focus on the emerging field of physics-informed machine learning (ML) quantum photonics that is envisioned to play a decisive role in addressing the above challenges. Specifically, three directions of ML-assisted quantum research are discussed: (i) rapid preselection of single single-photon sources via ML-assisted quantum measurements, (ii) hybrid ML-optimization approach for developing efficient quantum circuits elements, and (iii) ML-based frameworks for developing novel deterministic assembly of on-chip quantum emitters.

ACS Photonics•Dec 24, 2020•64 citations

Adaptive Quantum Readout Error Mitigation with Transfer Learning

Usama Inam Paracha, Syed Muhammad Abuzar Rizvi, Muhammad Mustafa Umar Gondel +2 more

Quantum computers, while offering exponential computational advantages, are highly vulnerable to noise-induced errors, particularly during the measurement process. This paper investigates machine learning-based error mitigation strategies with a focus on transfer learning. By initially training a base model on a specific quantum device, we leverage transfer learning to adapt the model for use on different quantum computers. This approach not only achieves superior performance with reduced dataset size and training time but also maintains hardware agnosticism, as demonstrated through tests on various IBM quantum backends. Our method provides a practical solution for enhancing the accuracy and efficiency of noisy intermediate-scale quantum (NISQ) algorithms by improving machine learning-based error mitigation.

International Conference on Information and Communication Technology Convergence (ICTC)•Oct 16, 2024•0 citations
Abstract The use of single-photon sources (SPSs) is central to numerous systems and devices proposed amidst a modern surge in quantum technology. However, manufacturing schemes remain imperfect, and single-photon emission purity must often be experimentally verified via interferometry. Such a process is typically slow and costly, which has motivated growing research into whether SPS quality can be more rapidly inferred from incomplete emission statistics. Hence, this study is a sequel to previous work that demonstrated significant uncertainty in the standard method of quality estimation, i.e. the least-squares fitting of a physically motivated function, and asks: can machine learning (ML) do better? The study leverages eight datasets obtained from measurements involving an exemplary quantum emitter, i.e. a single InGaAs/GaAs epitaxial quantum dot; these eight contexts predominantly vary in the intensity of the exciting laser. Specifically, via a form of ‘transfer learning’, five ML models, three linear and two ensemble-based, are trained on data from seven of the contexts and tested on the eighth. Validation metrics quickly reveal that even a linear regressor can outperform standard fitting when it is tested on the same contexts it was trained on, but the success of transfer learning is less assured, even though statistical analysis, made possible by data augmentation, suggests its superiority as an early estimator. Accordingly, the study concludes by discussing future strategies for grappling with the problem of SPS context dissimilarity, e.g. feature engineering and model adaptation.
Abstract Novel methods for rapidly estimating single-photon source (SPS) quality have been promoted in recent literature to address the expensive and time-consuming nature of experimental validation via intensity interferometry. However, the frequent lack of uncertainty discussions and reproducible details raises concerns about their reliability. This study investigates the use of data augmentation, a machine learning technique, to supplement experimental data with bootstrapped samples and quantify the uncertainty of such estimates. Eight datasets obtained from measurements involving a single InGaAs/GaAs epitaxial quantum dot serve as a proof-of-principle example. Analysis of one of the SPS quality metrics derived from efficient histogram fitting of the synthetic samples, i.e. the probability of multi-photon emission events, reveals significant uncertainty contributed by stochastic variability in the Poisson processes that describe detection rates. Ignoring this source of error risks severe overconfidence in both early quality estimates and claims for state-of-the-art SPS devices. Additionally, this study finds that standard least-squares fitting is comparable to using a Poisson likelihood, and expanding averages show some promise for early estimation. Also, reducing background counts improves fitting accuracy but does not address the Poisson-process variability. Ultimately, data augmentation demonstrates its value in supplementing physical experiments; its benefit here is to emphasise the need for a cautious assessment of SPS quality.
With wide applications ranging from quantum communication and metrology to biomedicine, single photon sources in solid-state hosts have become a major area of study. Here, we focus on three applications: nanothermometry, optically detected magnetic resonance (ODMR), and second order autocorrelation. We present novel statistical and machine learning (ML) approaches to extract information from experimental and simulated data and benchmark these methods against traditional inference-based statistical approaches. We found that compared to traditional inference-based methods ML algorithms can: i) predict temperatures at the nanoscale with greater accuracy and with less calibration points than traditional fitting methods; ii) identify the resonance peaks in ODMR spectra with factors ~1.3x and ~4.7x better accuracy and resolution and achieved equal or better performance with ~5x less data; and iii) have the potential to parse second order autocorrelation data more efficiently. ML algorithms are thus powerful tools for quantum sensing techniques. Author Keywords: colour centers, machine learning, nanosensing, nanothermometry, optically detected magnetic resonance, second order autocorrelation
Model-independent estimation of the properties of quantum states is a central challenge in quantum technologies, as experimental imperfections, drifts, and imprecise models of the actual quantum dynamics inevitably hinder accurate reconstructions. Here, we introduce a training strategy for photonic quantum extreme learning machines in which both the learning stage and the optimization of the measurement settings are performed entirely with classical light, while inference is carried out on genuinely quantum states. The protocol is based on the identity between the normalized output intensities following the evolution of coherent states through a linear optical reservoir, and the output statistics obtained with separable input quantum states. Building on this correspondence, we implemented a model-free, gradient-based optimization of the reservoir measurement projection directly on experimental data, without relying on a prior model of the device transformation. We experimentally show that the resulting classical-to-quantum transfer enables accurate reconstruction of single-qubit Pauli observables for previously unseen single-photon states, and extends to the estimation of a two-qubit entanglement witness for arbitrary bipartite states. Beyond demonstrating a qualitatively distinct form of out-of-distribution generalization across the classical-to-quantum boundary, our results identify a practical route to fast, adaptive, and resource-efficient training of photonic quantum learning devices.
Characterization of the photon statistics of an optical field is central to identifying light sources in various quantum-enabled technologies, from super-resolved quantum imaging to precise quantum sensing and spectroscopy. In practice, however, estimating photon-correlation functions with sufficient accuracy requires long acquisition times and computationally intensive statistical estimators, severely limiting real-time operation and the study of rapidly varying or weak sources. Here, we show that learned inference based on artificial neural networks enables rapid estimation of second-order coherence directly from photon-number-resolved measurements, outperforming the precision of conventional statistical methods with processing times three orders of magnitude shorter. Our approach is source-agnostic and remains accurate in few-photon illumination regimes where the statistics of coherent, partially coherent, thermal, and super-thermal light become nearly indistinguishable. We experimentally validate the method across more than 4000 distinct classical light sources, demonstrating equal or improved precision with drastically reduced data and computational requirements. These results establish machine-learning-enabled photon-statistics inference as a new operational route to rapid optical-coherence diagnostics, enabling real-time monitoring of classical light sources in regimes previously inaccessible to conventional estimators.
Abstract Realization of integrated quantum photonics is a key step toward scalable quantum applications such as quantum computing, sensing, information processing, and quantum material metrology. To enable practical quantum photonic systems, several challenges should be addressed, including (i) the realization of deterministic, bright, and stable single-photon emission operating at THz rates and at room temperatures, (ii) on-chip integration of efficient single-photon sources, and (iii) the development of deterministic and scalable nanoassembly of quantum circuitry elements. In this Perspective, we focus on the emerging field of physics-informed machine learning (ML) quantum photonics that is envisioned to play a decisive role in addressing the above challenges. Specifically, three directions of ML-assisted quantum research are discussed: (i) rapid preselection of single single-photon sources via ML-assisted quantum measurements, (ii) hybrid ML-optimization approach for developing efficient quantum circuits elements, and (iii) ML-based frameworks for developing novel deterministic assembly of on-chip quantum emitters.
Quantum computers, while offering exponential computational advantages, are highly vulnerable to noise-induced errors, particularly during the measurement process. This paper investigates machine learning-based error mitigation strategies with a focus on transfer learning. By initially training a base model on a specific quantum device, we leverage transfer learning to adapt the model for use on different quantum computers. This approach not only achieves superior performance with reduced dataset size and training time but also maintains hardware agnosticism, as demonstrated through tests on various IBM quantum backends. Our method provides a practical solution for enhancing the accuracy and efficiency of noisy intermediate-scale quantum (NISQ) algorithms by improving machine learning-based error mitigation.