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:2601.17465

Bayesian quantum sensing using graybox machine learning

Akram Youssry, Stefan Todd, Patrick Murton, Muhammad Junaid Arshad, Nicholas Werren, Alberto Peruzzo, Cristian Bonato

quant-phcs.LGeess.SPSubmitted Jan 24, 2026• Announced Jan 27, 2026• 1 citation• Revised Sep 10, 2026 (v2)
v1 · Jan 24, 2026v2 · Sep 10, 2026
arXivHTMLPDF

Abstract

Quantum sensors offer significant advantages over classical devices in spatial resolution and sensitivity, enabling transformative applications across materials science, healthcare, and beyond. Their practical performance, however, is often constrained by unmodelled effects, including noise, imperfect state preparation, and non-ideal control fields. In this work, we report the first experimental implementation of a graybox modelling strategy for a solid-state open quantum system. The graybox framework integrates a physics-based system model with a data-driven description of experimental imperfections, achieving higher fidelity than purely analytical (whitebox) approaches while requiring fewer training resources than fully deep-learning blackbox models. We experimentally validate the method on the task of estimating a static magnetic field using a single-spin quantum sensor, performing Bayesian inference with a graybox model trained on prior experimental data. Using roughly 10,000 training datapoints, the graybox model yields several orders of magnitude improvement in mean squared error over the corresponding physics-only model, as well as significant improvement over a blackbox model with comparable size. These results are broadly applicable to a wide range of quantum sensing platforms, not limited to single-spin systems, and are particularly valuable for real-time adaptive protocols, where model inaccuracies can otherwise lead to suboptimal control and degraded performance.

Related papers

Probabilistic Graybox Characterization of Quantum Devices With Bayesian Neural Networks

Poramet Pathumsoot, Michal Hajdušek, Rodney Van Meter

While the Graybox characterization method allows for implicit noise models and is platform-agnostic, the method lacks uncertainty quantification. Characterization of quantum devices is a crucial process that enables researchers to gain insight from experimental settings. Graybox characterization combines known system dynamics with unknown transformations, where the latter is modeled using machine learning. Prediction uncertainty helps researchers make informed decisions. It allows valuable insights from the devices without overconfidence. We therefore develop a probabilistic Graybox characterization model using probabilistic machine learning, specifically Bayesian Neural Networks, and utilize binary measurement outcomes directly for inference. With stochastic noise in a quantum device, we analyze statistical properties of the measurement data. Our results show that the model's prediction performance solely depends on its ability to capture the expected value of the true expectation value. Our proposed probabilistic Graybox model outperforms the original model by up to 1.9 times in capturing the distribution of observed data. We expect that our results will serve as an additional tool for characterizing quantum devices with uncertainty estimation, as they provide a flexible choice that can be utilized even without extensive prior knowledge of the noise model of the devices.

arXiv:2509.24232•IEEE Transactions on Quantum Engineering•Jan 1, 2026•2 citations

Deep reinforcement learning for quantum multiparameter estimation

Valeria Cimini, Mauro Valeri, Emanuele Polino +6 more

Estimation of physical quantities is at the core of most scientific research, and the use of quantum devices promises to enhance its performances. In real scenarios, it is fundamental to consider that resources are limited, and Bayesian adaptive estimation represents a powerful approach to efficiently allocate, during the estimation process, all the available resources. However, this framework relies on the precise knowledge of the system model, retrieved with a fine calibration, with results that are often computationally and experimentally demanding. We introduce a model-free and deep-learning-based approach to efficiently implement realistic Bayesian quantum metrology tasks accomplishing all the relevant challenges, without relying on any a priori knowledge of the system. To overcome this need, a neural network is trained directly on experimental data to learn the multiparameter Bayesian update. Then the system is set at its optimal working point through feedback provided by a reinforcement learning algorithm trained to reconstruct and enhance experiment heuristics of the investigated quantum sensor. Notably, we prove experimentally the achievement of higher estimation performances than standard methods, demonstrating the strength of the combination of these two black-box algorithms on an integrated photonic circuit. Our work represents an important step toward fully artificial intelligence-based quantum metrology.

arXiv:2209.00671•Advanced Photonics•Feb 6, 2023•58 citations

End-to-end Optimization of Single-Shot Quantum Machine Learning for Bayesian Inference

Theodoros Ilias, Fangjun Hu, Marti Vives +1 more

We introduce an end-to-end optimization strategy for quantum machine learning that directly targets performance under finite measurement resources, where learning objectives are defined directly at the level of task performance. The method is applied on a Bayesian quantum metrology task since it provides a natural testbed with known fundamental limits and scaling with system size. The sampling-aware hybrid algorithm achieves a single-shot risk within 1 dB of the -20 dB Bayesian limit using 32 qubits. We extend the Bayesian framework from parameter estimation to global function inference, where the task is to infer a target function of the sensor input drawn from an arbitrary prior, and we demonstrate a clear computational-sensing advantage for direct functional inference over indirect reconstruction. We relate the corresponding Bayesian risk to the Capacity metric and argue that the Resolvable Expressive Capacity provides a natural measure of the space of functions accessible in a single shot. The resulting eigentask analysis identifies noise-robust feature combinations that yield compact estimators with improved accuracy and reduced optimization cost in resource-limited or real-time on-device settings. Going beyond the irreducible statistical uncertainty associated with finite measurement records, we study implementation-induced distortions of the sensor response focusing on representative imperfections arising from readout errors and static coherent gate disorder. We show that when these imperfections are stationary across training and inference, end-to-end optimization can adapt the quantum circuit and classical estimator to the implemented hardware response, preserving high inference performance over a broad range of imperfection strengths.

quant-pharXiv:2512.20492•Dec 24, 2025•4 citations

Experimental graybox quantum system identification and control

Akram Youssry, Yang Yang, Robert J. Chapman +4 more

Abstract Understanding and controlling engineered quantum systems is key to developing practical quantum technology. However, given the current technological limitations, such as fabrication imperfections and environmental noise, this is not always possible. To address these issues, a great deal of theoretical and numerical methods for quantum system identification and control have been developed. These methods range from traditional curve fittings, which are limited by the accuracy of the model that describes the system, to machine learning (ML) methods, which provide efficient control solutions but no control beyond the output of the model, nor insights into the underlying physical process. Here we experimentally demonstrate a ‘graybox’ approach to construct a physical model of a quantum system and use it to design optimal control. We report superior performance over model fitting, while generating unitaries and Hamiltonians, which are quantities not available from the structure of standard supervised ML models. Our approach combines physics principles with high-accuracy ML and is effective with any problem where the required controlled quantities cannot be directly measured in experiments. This method naturally extends to time-dependent and open quantum systems, with applications in quantum noise spectroscopy and cancellation.

arXiv:2206.12201•npj Quantum Information•Jan 13, 2024•29 citations

Machine Learning for Quantum Sensing Systems: An Integrated Survey of Preparation, Measurement, Inference, and Control

Feng Li, Zhuo Wang, Derong Liu +9 more

Machine learning (ML) is increasingly embedded across quantum-sensing systems, yet the evidence remains fragmented across platforms, algorithms, and validation settings. This survey develops an integrated framework that organizes ML contributions into four coupled functions: preparation, measurement, inference, and control. It compares atomic, photonic, solid-state, superconducting, mechanical, and hybrid sensors using common criteria: physical bottleneck, data provenance, metrological objective, validation level, robustness, and deployment constraints. The analysis links recurring sensing structures to suitable learning paradigms and distinguishes improvements in quantum-resource preparation, information acquisition, parameter estimation, and system-level operation. Across the literature, inference has the broadest experimental coverage, while direct demonstrations also establish ML-assisted cold-atom preparation, adaptive measurement, and protocols coupling resource generation with online estimation. By contrast, evidence for sustained closed-loop control and autonomy remains device-specific, making stability, uncertainty calibration, transfer, latency, safety, and recovery central research priorities. Crucially, ML may reduce estimator risk, calibration burden, or operating cost, but it cannot recover information absent from a fixed measurement record. The resulting taxonomy and resource-matched attribution framework provide a consistent basis for comparing reported gains and designing reliable, transferable, and experimentally verifiable ML-enabled quantum sensors.

IEEE Transactions on Systems Man and Cybernetics Systems•Jan 1, 2026•0 citations

A Framework for Spatial Quantum Sensing

Luís Bugalho, Yasser Omar, Damian Markham

Quantum sensor networks promise precision advantages over classical and single-sensor strategies, in particular when the estimator is non-local. We address the problem of finding such estimators through a framework we connote spatial quantum sensing: given an underlying field interrogated by a network of quantum sensors at fixed positions, construct an estimator for a property of the field, for example, distinguishing a source of signal, or evaluating the field or its derivatives at an arbitrary point. We first treat polynomial fields, casting the task as an interpolation problem, and then generalize to fields modeled by analytic functions, which yields general least-squares estimators. A central and largely unaddressed question is under what conditions on sensor placement these estimators are well-defined and error-free. For $m$-dimensional arrays we give explicit constructions and proofs in the interpolation setting using algebraic geometry, and establish necessary and sufficient conditions in the general case. Comparing a non-local entangled protocol with the best local strategy, we show that entanglement yields maximal precision in distributed sensing under global resource constraints. Finally, we introduce error-free subspaces, a technique that translates prior knowledge of the field into a reduction in the number of required sensors. We expect these techniques to be broadly useful in sensing problems across scales, ranging from earth-scale experiments to local applications such as biological imaging.

quant-pharXiv:2602.12193•Feb 13, 2026•1 citation
Quantum sensors offer significant advantages over classical devices in spatial resolution and sensitivity, enabling transformative applications across materials science, healthcare, and beyond. Their practical performance, however, is often constrained by unmodelled effects, including noise, imperfect state preparation, and non-ideal control fields. In this work, we report the first experimental implementation of a graybox modelling strategy for a solid-state open quantum system. The graybox framework integrates a physics-based system model with a data-driven description of experimental imperfections, achieving higher fidelity than purely analytical (whitebox) approaches while requiring fewer training resources than fully deep-learning blackbox models. We experimentally validate the method on the task of estimating a static magnetic field using a single-spin quantum sensor, performing Bayesian inference with a graybox model trained on prior experimental data. Using roughly 10,000 training datapoints, the graybox model yields several orders of magnitude improvement in mean squared error over the corresponding physics-only model, as well as significant improvement over a blackbox model with comparable size. These results are broadly applicable to a wide range of quantum sensing platforms, not limited to single-spin systems, and are particularly valuable for real-time adaptive protocols, where model inaccuracies can otherwise lead to suboptimal control and degraded performance.
While the Graybox characterization method allows for implicit noise models and is platform-agnostic, the method lacks uncertainty quantification. Characterization of quantum devices is a crucial process that enables researchers to gain insight from experimental settings. Graybox characterization combines known system dynamics with unknown transformations, where the latter is modeled using machine learning. Prediction uncertainty helps researchers make informed decisions. It allows valuable insights from the devices without overconfidence. We therefore develop a probabilistic Graybox characterization model using probabilistic machine learning, specifically Bayesian Neural Networks, and utilize binary measurement outcomes directly for inference. With stochastic noise in a quantum device, we analyze statistical properties of the measurement data. Our results show that the model's prediction performance solely depends on its ability to capture the expected value of the true expectation value. Our proposed probabilistic Graybox model outperforms the original model by up to 1.9 times in capturing the distribution of observed data. We expect that our results will serve as an additional tool for characterizing quantum devices with uncertainty estimation, as they provide a flexible choice that can be utilized even without extensive prior knowledge of the noise model of the devices.
Estimation of physical quantities is at the core of most scientific research, and the use of quantum devices promises to enhance its performances. In real scenarios, it is fundamental to consider that resources are limited, and Bayesian adaptive estimation represents a powerful approach to efficiently allocate, during the estimation process, all the available resources. However, this framework relies on the precise knowledge of the system model, retrieved with a fine calibration, with results that are often computationally and experimentally demanding. We introduce a model-free and deep-learning-based approach to efficiently implement realistic Bayesian quantum metrology tasks accomplishing all the relevant challenges, without relying on any a priori knowledge of the system. To overcome this need, a neural network is trained directly on experimental data to learn the multiparameter Bayesian update. Then the system is set at its optimal working point through feedback provided by a reinforcement learning algorithm trained to reconstruct and enhance experiment heuristics of the investigated quantum sensor. Notably, we prove experimentally the achievement of higher estimation performances than standard methods, demonstrating the strength of the combination of these two black-box algorithms on an integrated photonic circuit. Our work represents an important step toward fully artificial intelligence-based quantum metrology.
We introduce an end-to-end optimization strategy for quantum machine learning that directly targets performance under finite measurement resources, where learning objectives are defined directly at the level of task performance. The method is applied on a Bayesian quantum metrology task since it provides a natural testbed with known fundamental limits and scaling with system size. The sampling-aware hybrid algorithm achieves a single-shot risk within 1 dB of the -20 dB Bayesian limit using 32 qubits. We extend the Bayesian framework from parameter estimation to global function inference, where the task is to infer a target function of the sensor input drawn from an arbitrary prior, and we demonstrate a clear computational-sensing advantage for direct functional inference over indirect reconstruction. We relate the corresponding Bayesian risk to the Capacity metric and argue that the Resolvable Expressive Capacity provides a natural measure of the space of functions accessible in a single shot. The resulting eigentask analysis identifies noise-robust feature combinations that yield compact estimators with improved accuracy and reduced optimization cost in resource-limited or real-time on-device settings. Going beyond the irreducible statistical uncertainty associated with finite measurement records, we study implementation-induced distortions of the sensor response focusing on representative imperfections arising from readout errors and static coherent gate disorder. We show that when these imperfections are stationary across training and inference, end-to-end optimization can adapt the quantum circuit and classical estimator to the implemented hardware response, preserving high inference performance over a broad range of imperfection strengths.
Abstract Understanding and controlling engineered quantum systems is key to developing practical quantum technology. However, given the current technological limitations, such as fabrication imperfections and environmental noise, this is not always possible. To address these issues, a great deal of theoretical and numerical methods for quantum system identification and control have been developed. These methods range from traditional curve fittings, which are limited by the accuracy of the model that describes the system, to machine learning (ML) methods, which provide efficient control solutions but no control beyond the output of the model, nor insights into the underlying physical process. Here we experimentally demonstrate a ‘graybox’ approach to construct a physical model of a quantum system and use it to design optimal control. We report superior performance over model fitting, while generating unitaries and Hamiltonians, which are quantities not available from the structure of standard supervised ML models. Our approach combines physics principles with high-accuracy ML and is effective with any problem where the required controlled quantities cannot be directly measured in experiments. This method naturally extends to time-dependent and open quantum systems, with applications in quantum noise spectroscopy and cancellation.
Machine learning (ML) is increasingly embedded across quantum-sensing systems, yet the evidence remains fragmented across platforms, algorithms, and validation settings. This survey develops an integrated framework that organizes ML contributions into four coupled functions: preparation, measurement, inference, and control. It compares atomic, photonic, solid-state, superconducting, mechanical, and hybrid sensors using common criteria: physical bottleneck, data provenance, metrological objective, validation level, robustness, and deployment constraints. The analysis links recurring sensing structures to suitable learning paradigms and distinguishes improvements in quantum-resource preparation, information acquisition, parameter estimation, and system-level operation. Across the literature, inference has the broadest experimental coverage, while direct demonstrations also establish ML-assisted cold-atom preparation, adaptive measurement, and protocols coupling resource generation with online estimation. By contrast, evidence for sustained closed-loop control and autonomy remains device-specific, making stability, uncertainty calibration, transfer, latency, safety, and recovery central research priorities. Crucially, ML may reduce estimator risk, calibration burden, or operating cost, but it cannot recover information absent from a fixed measurement record. The resulting taxonomy and resource-matched attribution framework provide a consistent basis for comparing reported gains and designing reliable, transferable, and experimentally verifiable ML-enabled quantum sensors.
Quantum sensor networks promise precision advantages over classical and single-sensor strategies, in particular when the estimator is non-local. We address the problem of finding such estimators through a framework we connote spatial quantum sensing: given an underlying field interrogated by a network of quantum sensors at fixed positions, construct an estimator for a property of the field, for example, distinguishing a source of signal, or evaluating the field or its derivatives at an arbitrary point. We first treat polynomial fields, casting the task as an interpolation problem, and then generalize to fields modeled by analytic functions, which yields general least-squares estimators. A central and largely unaddressed question is under what conditions on sensor placement these estimators are well-defined and error-free. For mmm-dimensional arrays we give explicit constructions and proofs in the interpolation setting using algebraic geometry, and establish necessary and sufficient conditions in the general case. Comparing a non-local entangled protocol with the best local strategy, we show that entanglement yields maximal precision in distributed sensing under global resource constraints. Finally, we introduce error-free subspaces, a technique that translates prior knowledge of the field into a reduction in the number of required sensors. We expect these techniques to be broadly useful in sensing problems across scales, ranging from earth-scale experiments to local applications such as biological imaging.