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

IEEE Transactions on Systems Man and Cybernetics Systems

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

Feng Li, Zhuo Wang, Derong Liu, Kai Wei, Shan Xue, Ruigang Wang, Xinxiu Zhou, Haoying Pang, Qi Zheng, Li jiahang, Xusheng Lei, Wei Quan

Published Jan 1, 2026• 0 citations
Publisher

Abstract

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.

DOI
10.1109/tsmc.2026.3744356

Related papers

A systematic analysis of data-driven signal processing for quantum sensing

Modou Njie

Quantum sensing offers sensitivities surpassing classical limits, yet practical deployment is often constrained by noise and inefficient signal extraction. This systematic review synthesizes literature from 2015 to 2025 and maps the landscape of data-driven signal processing across diverse platforms, including nitrogen-vacancy (NV) centers and cold atom interferometers. Emerging methodologies are categorized and emphasis is placed on the dominance of hybrid quantum-classical architectures where classical machine learning manages high-dimensional data processing while quantum processors execute specific primitives like feature mapping. Our findings demonstrate that advanced statistical frameworks such as Variational Bayesian Inference (VBI) enable scalable multi-parameter estimation with polynomial complexity, allowing dozens of sensor parameters to be estimated within minutes. Meanwhile, quantum sensing missions in GPS-denied navigation have achieved operational gains through AI-automated software designed to manage quantum hardware drifts and environmental noise, despite a persistent reality gap for generalized quantum machine learning. We further analyze the critical sensitivity-robustness dilemma inherent in entangled sensors and discuss mitigation strategies such as covariant quantum error-correcting codes. We also distinguish between theoretical Quantum Fisher Information (QFI) and realized Mean Squared Error (MSE) of sensor outputs to resolve benchmarking inconsistencies in quantum metrology, and propose a "Gold Standard" reproducibility framework tailored for the Noisy Intermediate-Scale Quantum (NISQ) era to promote automated workflows and hardware calibration transparency. This review provides a strategic roadmap for researchers to bridge the gap between theoretical sensitivity and reliable, software-defined operational utility in quantum sensing.

International Journal of Frontline Research in Engineering and Technology•Mar 20, 2026•0 citations

Machine Learning for Advancing Quantum Field Theory-Enabled Sensor Networks in Precision Disease Detection: A Review

Pratik Khanal, Amrit Khanal

Quantum field theory (QFT)-enabled sensor networks are revolutionizing precision disease detection by exploiting quantum–classical correlations to achieve sensitivity and specificity beyond classical limits. Their diagnostic power, however, depends critically on machine learning (ML) for denoising, fusion, and interpretation of high-dimensional, entangled, and multimodal data streams. This review integrates advances at the ML–QFT interface across four pillars: (i) sensing foundations—NV-diamond and optically pumped magnetometry, cavity optomechanics, and quantum plasmonic or photonic systems; (ii) learning methods—physics-informed preprocessing (e.g., VAEs, diffusion models, PINNs), representation learning for sensor arrays and time series (GNNs, transformers), and hybrid quantum–classical architectures; (iii) applications—ultrasensitive pathogen detection, cancer biomarker profiling, neurodegenerative disease monitoring, and epidemiological surveillance; and (iv) cross-cutting enablers—adaptive calibration, federated and transfer learning, and explainable AI for clinical assurance. A practical ML–QFT co-design framework is presented, mapping model classes to sensor physics and deployment settings (edge/on-sensor versus cloud). An evaluation checklist coupling metrological and ML metrics—limit of detection, SNR uplift, calibration stability, latency, robustness under data shift, and interpretability—is proposed for benchmarking. Literature evidence shows that physics-aligned representations and hybrid learners consistently enhance performance in low-SNR and data-scarce regimes, though challenges remain from decoherence, drift, and scalability. The review concludes with a roadmap toward open benchmarks, quantum-networked distributed sensing, multiscale modeling linking molecular and population data, and certifiable, explainable ML pipelines—positioning ML not just as post-hoc analytics but as a design logic for globally distributed, self-calibrating quantum diagnostic ecosystems.

Crimson Journal of Science and Technology•Dec 30, 2025•1 citation

Principles, Platforms and Real-World Applications of Quantum Sensing: A Review

Alyx Wilhelm, Nicolas Wanner, Archana T. Anilkumar +3 more

Quantum sensing converts perturbations of controlled quantum states into estimates of magnetic and electric fields, temperature, time and frequency, acceleration, rotation, gravity, and local spin environments. Although the field has produced exceptional laboratory demonstrations, comparisons across platforms are often obscured by incompatible metrics and by application-led descriptions that do not connect quantum dynamics to complete sensor architectures. This review develops a unified, sensor-centered framework linking Hamiltonian parameter encoding, superposition, coherence, decoherence, entanglement, quantum Fisher information, and readout to experimentally reported performance. Nitrogen-vacancy centers in diamond, optically pumped magnetometers, atomic and optical clocks, atom-interferometric inertial and gravity sensors, and Rydberg-atom electric-field sensors are examined using common criteria that include sensitivity, minimum detectable signal, response time, bandwidth, selectivity, spatial resolution, stability, dynamic range, operating conditions, calibration, and deployment maturity. Photonic quantum enhancement, quantum illumination, molecular-qubit sensing, and selected emerging platforms are discussed as cross-cutting cases. The review retains application coverage in biomedicine, navigation, timing, geophysics, agriculture, environmental monitoring, and defense, while distinguishing coherent quantum sensors from quantum-material-enabled conventional sensors. The comparison shows that practical advantage is rarely determined by sensitivity alone; sensor-to-source distance, shielding, laser and vacuum infrastructure, coherence, readout efficiency, calibration, packaging, environmental robustness, and cost frequently determine application viability. A deployment-oriented roadmap is presented for integrated control and readout, standardized benchmarking, hybrid quantum-classical operation, and responsible translation.

IEEE Sensors Journal•Jan 1, 2026•0 citations

Quantum-Enabled Indoor Positioning: A Survey on Sensing, Computing, and Learning Approaches

Yonghao Yu, Youyang Qu, Yongwei Tang +6 more

Indoor positioning is essential for Internet of Things (IoT) applications, mobile robots, smart buildings, and industrial automation. Although classical positioning technologies have achieved considerable maturity, challenges remain in complex propagation environments, high-precision measurements, computational efficiency, and system robustness. This survey critically examines the potential of quantum technologies to complement existing indoor positioning systems through quantum sensing, quantum optimization, and quantum machine learning (QML). An error-budget perspective is introduced to distinguish measurement uncertainty from propagation effects, geometric limitations, device variations, and environmental changes, thereby clarifying when quantum-enhanced measurements may improve positioning accuracy. Existing studies are systematically compared in terms of technical mechanisms, evidence maturity, computational requirements, and practical limitations, while distinguishing physical quantum implementations from simulation-based and quantum-inspired approaches. Particular attention is given to localization accuracy, latency, energy consumption, hardware requirements, and robustness under realistic operating conditions. The analysis indicates that current quantum-assisted positioning research remains largely exploratory, with more practical near-term opportunities in offline optimization, calibration, and hybrid processing than in real-time positioning. Finally, this survey identifies key research challenges and proposes a prioritized roadmap with evaluation criteria for future quantum-assisted indoor positioning systems.

IEEE Internet of Things Journal•Jan 1, 2026•0 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

A Survey of Quantum Machine Learning: Foundations, Algorithms, Frameworks, Data and Applications

Francesc Rodríguez-Díaz, David Gutiérrez‐Avilés, Alicia Troncoso +1 more

Quantum machine learning combines quantum computing with machine learning to solve complex computational problems more efficiently than classical approaches. This survey provides an introduction to the foundations, algorithms, frameworks, data and applications of quantum machine learning, serving as a resource for researchers and practitioners. We begin by reviewing existing surveys to identify gaps that this work addresses, followed by a detailed discussion of the foundational principles of quantum mechanics and machine learning essential for quantum machine learning. Key algorithms are examined, highlighting their mechanisms, advantages, and applications across various domains. Current frameworks and platforms for implementing quantum machine learning algorithms are explored, emphasizing their unique features and suitability for different contexts. Existing quantum datasets for practical usage are also reported and commented on. This survey also reviews over 135 articles, categorized into theoretical and practical contributions, to identify key advances, limitations, and application areas within quantum machine learning. Critical challenges such as hardware limitations, error rates, and scalability are analyzed to detect the obstacles that must be addressed for practical deployment. By synthesizing these elements into a structured overview, this survey aims at serving as both an introduction and a guide for advancing research and development in this disruptive field.

ACM Computing Surveys•Sep 11, 2025•46 citations
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 sensing offers sensitivities surpassing classical limits, yet practical deployment is often constrained by noise and inefficient signal extraction. This systematic review synthesizes literature from 2015 to 2025 and maps the landscape of data-driven signal processing across diverse platforms, including nitrogen-vacancy (NV) centers and cold atom interferometers. Emerging methodologies are categorized and emphasis is placed on the dominance of hybrid quantum-classical architectures where classical machine learning manages high-dimensional data processing while quantum processors execute specific primitives like feature mapping. Our findings demonstrate that advanced statistical frameworks such as Variational Bayesian Inference (VBI) enable scalable multi-parameter estimation with polynomial complexity, allowing dozens of sensor parameters to be estimated within minutes. Meanwhile, quantum sensing missions in GPS-denied navigation have achieved operational gains through AI-automated software designed to manage quantum hardware drifts and environmental noise, despite a persistent reality gap for generalized quantum machine learning. We further analyze the critical sensitivity-robustness dilemma inherent in entangled sensors and discuss mitigation strategies such as covariant quantum error-correcting codes. We also distinguish between theoretical Quantum Fisher Information (QFI) and realized Mean Squared Error (MSE) of sensor outputs to resolve benchmarking inconsistencies in quantum metrology, and propose a "Gold Standard" reproducibility framework tailored for the Noisy Intermediate-Scale Quantum (NISQ) era to promote automated workflows and hardware calibration transparency. This review provides a strategic roadmap for researchers to bridge the gap between theoretical sensitivity and reliable, software-defined operational utility in quantum sensing.
Quantum field theory (QFT)-enabled sensor networks are revolutionizing precision disease detection by exploiting quantum–classical correlations to achieve sensitivity and specificity beyond classical limits. Their diagnostic power, however, depends critically on machine learning (ML) for denoising, fusion, and interpretation of high-dimensional, entangled, and multimodal data streams. This review integrates advances at the ML–QFT interface across four pillars: (i) sensing foundations—NV-diamond and optically pumped magnetometry, cavity optomechanics, and quantum plasmonic or photonic systems; (ii) learning methods—physics-informed preprocessing (e.g., VAEs, diffusion models, PINNs), representation learning for sensor arrays and time series (GNNs, transformers), and hybrid quantum–classical architectures; (iii) applications—ultrasensitive pathogen detection, cancer biomarker profiling, neurodegenerative disease monitoring, and epidemiological surveillance; and (iv) cross-cutting enablers—adaptive calibration, federated and transfer learning, and explainable AI for clinical assurance. A practical ML–QFT co-design framework is presented, mapping model classes to sensor physics and deployment settings (edge/on-sensor versus cloud). An evaluation checklist coupling metrological and ML metrics—limit of detection, SNR uplift, calibration stability, latency, robustness under data shift, and interpretability—is proposed for benchmarking. Literature evidence shows that physics-aligned representations and hybrid learners consistently enhance performance in low-SNR and data-scarce regimes, though challenges remain from decoherence, drift, and scalability. The review concludes with a roadmap toward open benchmarks, quantum-networked distributed sensing, multiscale modeling linking molecular and population data, and certifiable, explainable ML pipelines—positioning ML not just as post-hoc analytics but as a design logic for globally distributed, self-calibrating quantum diagnostic ecosystems.
Quantum sensing converts perturbations of controlled quantum states into estimates of magnetic and electric fields, temperature, time and frequency, acceleration, rotation, gravity, and local spin environments. Although the field has produced exceptional laboratory demonstrations, comparisons across platforms are often obscured by incompatible metrics and by application-led descriptions that do not connect quantum dynamics to complete sensor architectures. This review develops a unified, sensor-centered framework linking Hamiltonian parameter encoding, superposition, coherence, decoherence, entanglement, quantum Fisher information, and readout to experimentally reported performance. Nitrogen-vacancy centers in diamond, optically pumped magnetometers, atomic and optical clocks, atom-interferometric inertial and gravity sensors, and Rydberg-atom electric-field sensors are examined using common criteria that include sensitivity, minimum detectable signal, response time, bandwidth, selectivity, spatial resolution, stability, dynamic range, operating conditions, calibration, and deployment maturity. Photonic quantum enhancement, quantum illumination, molecular-qubit sensing, and selected emerging platforms are discussed as cross-cutting cases. The review retains application coverage in biomedicine, navigation, timing, geophysics, agriculture, environmental monitoring, and defense, while distinguishing coherent quantum sensors from quantum-material-enabled conventional sensors. The comparison shows that practical advantage is rarely determined by sensitivity alone; sensor-to-source distance, shielding, laser and vacuum infrastructure, coherence, readout efficiency, calibration, packaging, environmental robustness, and cost frequently determine application viability. A deployment-oriented roadmap is presented for integrated control and readout, standardized benchmarking, hybrid quantum-classical operation, and responsible translation.
Indoor positioning is essential for Internet of Things (IoT) applications, mobile robots, smart buildings, and industrial automation. Although classical positioning technologies have achieved considerable maturity, challenges remain in complex propagation environments, high-precision measurements, computational efficiency, and system robustness. This survey critically examines the potential of quantum technologies to complement existing indoor positioning systems through quantum sensing, quantum optimization, and quantum machine learning (QML). An error-budget perspective is introduced to distinguish measurement uncertainty from propagation effects, geometric limitations, device variations, and environmental changes, thereby clarifying when quantum-enhanced measurements may improve positioning accuracy. Existing studies are systematically compared in terms of technical mechanisms, evidence maturity, computational requirements, and practical limitations, while distinguishing physical quantum implementations from simulation-based and quantum-inspired approaches. Particular attention is given to localization accuracy, latency, energy consumption, hardware requirements, and robustness under realistic operating conditions. The analysis indicates that current quantum-assisted positioning research remains largely exploratory, with more practical near-term opportunities in offline optimization, calibration, and hybrid processing than in real-time positioning. Finally, this survey identifies key research challenges and proposes a prioritized roadmap with evaluation criteria for future quantum-assisted indoor positioning systems.
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.
Quantum machine learning combines quantum computing with machine learning to solve complex computational problems more efficiently than classical approaches. This survey provides an introduction to the foundations, algorithms, frameworks, data and applications of quantum machine learning, serving as a resource for researchers and practitioners. We begin by reviewing existing surveys to identify gaps that this work addresses, followed by a detailed discussion of the foundational principles of quantum mechanics and machine learning essential for quantum machine learning. Key algorithms are examined, highlighting their mechanisms, advantages, and applications across various domains. Current frameworks and platforms for implementing quantum machine learning algorithms are explored, emphasizing their unique features and suitability for different contexts. Existing quantum datasets for practical usage are also reported and commented on. This survey also reviews over 135 articles, categorized into theoretical and practical contributions, to identify key advances, limitations, and application areas within quantum machine learning. Critical challenges such as hardware limitations, error rates, and scalability are analyzed to detect the obstacles that must be addressed for practical deployment. By synthesizing these elements into a structured overview, this survey aims at serving as both an introduction and a guide for advancing research and development in this disruptive field.