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:2610.40848

Predictive Maintenance at the Edge: An Embedded Anomaly Detection Pipeline Trained with Quantum Optimization

Filipe Chagas Ferraz

quant-phcs.LGSubmitted Oct 29, 2026• Announced Nov 2, 2026• 0 citations
arXivHTMLPDF

Abstract

This article presents a predictive-maintenance pipeline that detects anomalies in bearings from vibration signals, adapted for deployment at the edge on low-cost microcontrollers, and that uses quantum optimization in two steps of its training. The pipeline consists of a step of vibration feature extraction and a step of classification with a one-class support vector machine (OCSVM). The features are selected by minimum redundancy and maximum relevance (mRMR) with the variational quantum eigensolver (VQE), under a limit on computing time. The OCSVM is adapted for inference at the edge without floating-point operations and is trained by a variational quantum algorithm with Pauli-correlation encoding (PCE). In small-scale tests on a simulator, with the Paderborn bearing dataset, the VQE found the optimal selection in 36% of the runs with Adam, and PCE chose better support vectors than chance, until a classical local search made the two equal. A classical baseline, the usual OCSVM trained by SMO with its weights rounded, matched them without any search.

AI Summary

The paper studies whether quantum optimization can train a bearing-anomaly detector designed for integer-only inference on low-cost microcontrollers. The pipeline extracts vibration features and applies a one-class SVM with a base-2 Laplacian kernel and binary support-vector weights. Two training tasks are formulated as QUBOs. Feature selection chooses five of 16 features under an estimated cycle budget, using a minimum-redundancy–maximum-relevance objective. A 17-qubit VQE preserves the number of selected features and represents the shared FFT cost with an auxiliary variable. SVM training selects 32 of 320 healthy training windows using Pauli-correlation encoding (PCE) on 16 qubits. An adaptive offset enforces $\sum_i\tilde{x}_i=32$ in the continuous surrogate, avoiding difficulties with sparse-cardinality penalties. Experiments use simulated circuits and three bearing-level train/test splits of the Paderborn dataset. VQE with exact-gradient Adam finds the optimal feature subset in 36% of runs, checked against exhaustive enumeration of 4,368 subsets. PCE improves support-vector selection over random initialization before classical refinement, but local search removes the observed advantage. A conventional SMO-trained SVM with rounded weights matches the resulting detectors without local search. Better feature-selection objective values do not always yield better detection. The study does not demonstrate quantum advantage or hardware deployment. Stated limitations include one operating condition, six healthy bearings from one manufacturer, simulated quantum circuits, and analytically estimated rather than measured microcontroller costs. Feature relevance and decision thresholds use labeled artificial-damage data; fully unsupervised selection and calibration remain unresolved.

No sign of AI-generated textIncrementalSmall numerics onlyUnclear writing

Error found: The abstract's 36% optimal-selection rate is not independently supported by the supplied results section, whose corresponding rate and run count remain unresolved placeholders. The reported equivalence of detector performance after local search and with SMO plus rounding lacks inspectable numerical support in the supplied text and figures. The claimed equivalence between the displayed binary OCSVM objective's continuous relaxation and conventional convex OCSVM training is incorrect because dropping the diagonal term preserves the binary problem but changes the continuous problem.

Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.

Related papers

Unsupervised Quantum Anomaly Detection on Noisy Quantum Processors

Daniel Pranjić, Florian Knäble, Philipp Kunst +6 more

Whether in fundamental physics, cybersecurity or finance, the detection of anomalies with machine learning techniques is a highly relevant and active field of research, as it potentially accelerates the discovery of novel physics or criminal activities. We provide a systematic analysis of the generalization properties of the One-Class Support Vector Machine (OCSVM) algorithm, using projected quantum kernels for a realistic dataset of the latter application. These results were both theoretically simulated and experimentally validated on trapped-ion and superconducting quantum processors, by leveraging partial state tomography to obtain precise approximations of the quantum states that are used to estimate the quantum kernels. Moreover, we analyzed both platforms respective hardware-efficient feature maps over a wide range of anomaly ratios and showed that for our financial dataset in all anomaly regimes, the quantum-enhanced OCSVMs lead to better generalization properties compared to the purely classical approach. As such our work bridges the gap between theory and practice in the noisy intermediate scale quantum (NISQ) era and paves the path towards useful quantum applications.

arXiv:2411.16970•Nov 25, 2024•5 citations

Unsupervised Quantum Anomaly Detection on Noisy Quantum Processors

Daniel Pranjić, Philipp Kunst, Florian Knäble +6 more

Abstract Whether in fundamental physics, cybersecurity or finance, the detection of anomalies with machine learning techniques is a highly relevant and active field of research, as it potentially accelerates the discovery of novel physics or criminal activities. We provide a systematic analysis of the generalization properties of the One-Class Support Vector Machine (OCSVM) algorithm, using projected quantum kernels for a realistic dataset of the latter application. These results were both theoretically simulated and experimentally validated on trapped-ion and superconducting quantum processors, by leveraging partial state tomography to obtain precise approximations of the quantum states that are used to estimate the quantum kernels. Moreover, we analyzed both platforms respective hardware-efficient feature maps over a wide range of anomaly ratios and showed that for our financial dataset in all anomaly regimes, the quantum-enhanced OCSVMs lead to better generalization properties compared to the purely classical approach. As such our work bridges the gap between theory and practice in the noisy intermediate scale quantum (NISQ) era and paves the path towards useful quantum applications.

•Apr 21, 2025•2 citations

Quantum-Enhanced AI for Predictive Analytics in Manufacturing

Priti Sharma, Surendra Kumar, Mohammed H. Fallah +4 more

The contemporary manufacturing landscape is more complicated, data-heavy and more automated than ever before, producing enormous volumes of multi-modal sensor data, production logs and maintenance records. Predictive analytics within these environments becomes essential to achieve operational efficiency, reduce downtime, and optimise resource allocation; however, many classical machine learning methods struggle with high-dimensional, nonlinear, and dynamic data, resulting in suboptimal predictions and slow decision-making. To overcome these issues, the proposed research suggests a Quantum-Enhanced AI (QAI) system combining various quantum algorithms: Quantum Neural Networks (Quantum Neural Network) to recognize patterns, Quantum Supporter Vector Machines (Quantum Supporter Vector Machines) to classify, Quantum Approximate Optimization Algorithm (Quantum Approximate Optimization Algorithm) to schedule production and allocate resources, and Variational Quantum Eigensolver (Variational Quantum Eigensolver) to optimize parameters under noisy settings or conditions into a hybrid quantum-class The framework is a quantum computer based framework that encodes classical manufacturing data in the form of quantum states, using amplitude and qubit-efficient embeddings, allowing representation of high-dimensional features with the parallelism and entanglement of quantum computing. Realworld experimental validation of both turbofan engine degradation and semiconductor manufacturing sensor logs was conducted using IBM Qiskit and PennyLane to ensure reproducibility, with noise-aware training and error mitigation policies applied in simulations on NISQ-era hardware. The proposed QAI framework demonstrated higher predictive accuracy (12 to 15) than state-of-the-art hybrid quantum-classical and classical models, with superior accuracy, memorability, and F1 scores, and showed exceptional capabilities in anomaly detection, failure prediction, and optimisation. The results validate that quantum algorithms can be integrated into a predictive analytics pipeline to overcome classical computation bottlenecks, enhance robustness, and provide actionable insights for making real-time manufacturing decisions. Overall, this paper has demonstrated that quantum-assisted AI can be beneficial in intelligent manufacturing environments, laying the groundwork for expanding practical applications that bridge the gap between the promises of quantum computational power and predictive analytics in the industry.

IEEE International Conference on Green Industrial Electronics and Sustainable Technologies (GIEST)•Oct 11, 2025•0 citations

Semisupervised Anomaly Detection using Support Vector Regression with Quantum Kernel

Kilian Tscharke, Sebastian Issel, Pascal Debus

Anomaly detection, an important task in data analysis, involves identifying observations or events that deviate in some way from the rest of the data. Machine learning techniques have shown success in automating this process by detecting hidden patterns and deviations in large-scale data. The potential of quantum computing for machine learning has been widely recognized, leading to extensive research efforts to develop suitable quantum machine learning (QML) algorithms. In particular, the search for QML algorithms for near-term NISQ devices is in full swing. However, NISQ devices pose additional challenges due to their limited qubit coherence times, low number of qubits, and high error rates. Kernel methods based on quantum kernel estimation have emerged as a promising approach to QML on NISQ devices, offering theoretical guarantees, versatility, and compatibility with NISQ constraints. Especially support vector machines (SVM) utilizing quantum kernel estimation have shown success in various supervised learning tasks. However, in the context of anomaly detection, semisupervised learning is of great relevance, and yet there is limited research published in this area. This paper introduces an approach to semisupervised anomaly detection based on the reconstruction loss of a support vector regression (SVR) with quantum kernel. This novel model is an alternative to the variational quantum and quantum kernel one-class classifiers, and is compared to a quantum autoencoder as quantum baseline and a SVR with radial-basis-function (RBF) kernel as well as a classical autoencoder as classical baselines. The models are benchmarked extensively on 10 real-world anomaly detection data sets and one toy data set and it is shown that our SVR model with quantum kernel performs better than the SVR with RBF kernel as well as all other models, achieving highest mean AUC over all data sets. In addition, our QSVR outperforms the Quantum autoencoder on 9 out of 11 data sets.

arXiv:2308.00583•IEEE International Conference on Quantum Computing and Engineering (QCE)•Sep 17, 2023•24 citations

Optimizing Feature Selection for Predictive Analytics With Quantum Algorithms

C. N. Manushree, Hari Deepthi Putluru, Himanshu Khajuria +2 more

Quantum algorithms for predictive analytics is important across various sectors such as finance, healthcare, marketing and supply chain management where accurate predictions can be used to optimise operations and decision-making processes. This chapter addresses Specific quantum machine learning algorithms such as Quantum Support Vector Machines (QSVM), Quantum K-nearest Neighbors (QKNN) and Quantum Neural Networks (QNN). These algorithms enhance classification, regression and complex predictive tasks by accelerating training processes and managing large feature spaces effectively. And, to this, quantum optimization algorithms like Quantum Approximate Optimization Algorithm (QAOA) and Variational Quantum Eigensolver (VQE) are mentioned for their role in optimising model parameters and improving prediction accuracy. In addition, it discusses quantum data encoding methods, such as Quantum Fourier Transform (QFT) and amplitude encoding. Along with this, it acknowledges significant challenges such as hardware limitations, algorithmic complexity and practical integration issues.

Advances in marketing, customer relationship management, and e-services book series•Dec 13, 2024•5 citations

Variational Quantum Feature Selection for High-Dimensional Classification: A Hybrid Quantum-Classical Approach

Sharath Kumar Jagannathan, Thomas Abraham JV, Yogesh C +1 more

Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.

•Mar 23, 2026•0 citations
This article presents a predictive-maintenance pipeline that detects anomalies in bearings from vibration signals, adapted for deployment at the edge on low-cost microcontrollers, and that uses quantum optimization in two steps of its training. The pipeline consists of a step of vibration feature extraction and a step of classification with a one-class support vector machine (OCSVM). The features are selected by minimum redundancy and maximum relevance (mRMR) with the variational quantum eigensolver (VQE), under a limit on computing time. The OCSVM is adapted for inference at the edge without floating-point operations and is trained by a variational quantum algorithm with Pauli-correlation encoding (PCE). In small-scale tests on a simulator, with the Paderborn bearing dataset, the VQE found the optimal selection in 36% of the runs with Adam, and PCE chose better support vectors than chance, until a classical local search made the two equal. A classical baseline, the usual OCSVM trained by SMO with its weights rounded, matched them without any search.
The paper studies whether quantum optimization can train a bearing-anomaly detector designed for integer-only inference on low-cost microcontrollers. The pipeline extracts vibration features and applies a one-class SVM with a base-2 Laplacian kernel and binary support-vector weights.
Two training tasks are formulated as QUBOs. Feature selection chooses five of 16 features under an estimated cycle budget, using a minimum-redundancy–maximum-relevance objective. A 17-qubit VQE preserves the number of selected features and represents the shared FFT cost with an auxiliary variable. SVM training selects 32 of 320 healthy training windows using Pauli-correlation encoding (PCE) on 16 qubits. An adaptive offset enforces ∑ix~i=32\sum_i\tilde{x}_i=32∑i​x~i​=32 in the continuous surrogate, avoiding difficulties with sparse-cardinality penalties.
Experiments use simulated circuits and three bearing-level train/test splits of the Paderborn dataset. VQE with exact-gradient Adam finds the optimal feature subset in 36% of runs, checked against exhaustive enumeration of 4,368 subsets. PCE improves support-vector selection over random initialization before classical refinement, but local search removes the observed advantage. A conventional SMO-trained SVM with rounded weights matches the resulting detectors without local search. Better feature-selection objective values do not always yield better detection.
The study does not demonstrate quantum advantage or hardware deployment. Stated limitations include one operating condition, six healthy bearings from one manufacturer, simulated quantum circuits, and analytically estimated rather than measured microcontroller costs. Feature relevance and decision thresholds use labeled artificial-damage data; fully unsupervised selection and calibration remain unresolved.
Whether in fundamental physics, cybersecurity or finance, the detection of anomalies with machine learning techniques is a highly relevant and active field of research, as it potentially accelerates the discovery of novel physics or criminal activities. We provide a systematic analysis of the generalization properties of the One-Class Support Vector Machine (OCSVM) algorithm, using projected quantum kernels for a realistic dataset of the latter application. These results were both theoretically simulated and experimentally validated on trapped-ion and superconducting quantum processors, by leveraging partial state tomography to obtain precise approximations of the quantum states that are used to estimate the quantum kernels. Moreover, we analyzed both platforms respective hardware-efficient feature maps over a wide range of anomaly ratios and showed that for our financial dataset in all anomaly regimes, the quantum-enhanced OCSVMs lead to better generalization properties compared to the purely classical approach. As such our work bridges the gap between theory and practice in the noisy intermediate scale quantum (NISQ) era and paves the path towards useful quantum applications.
Abstract Whether in fundamental physics, cybersecurity or finance, the detection of anomalies with machine learning techniques is a highly relevant and active field of research, as it potentially accelerates the discovery of novel physics or criminal activities. We provide a systematic analysis of the generalization properties of the One-Class Support Vector Machine (OCSVM) algorithm, using projected quantum kernels for a realistic dataset of the latter application. These results were both theoretically simulated and experimentally validated on trapped-ion and superconducting quantum processors, by leveraging partial state tomography to obtain precise approximations of the quantum states that are used to estimate the quantum kernels. Moreover, we analyzed both platforms respective hardware-efficient feature maps over a wide range of anomaly ratios and showed that for our financial dataset in all anomaly regimes, the quantum-enhanced OCSVMs lead to better generalization properties compared to the purely classical approach. As such our work bridges the gap between theory and practice in the noisy intermediate scale quantum (NISQ) era and paves the path towards useful quantum applications.
The contemporary manufacturing landscape is more complicated, data-heavy and more automated than ever before, producing enormous volumes of multi-modal sensor data, production logs and maintenance records. Predictive analytics within these environments becomes essential to achieve operational efficiency, reduce downtime, and optimise resource allocation; however, many classical machine learning methods struggle with high-dimensional, nonlinear, and dynamic data, resulting in suboptimal predictions and slow decision-making. To overcome these issues, the proposed research suggests a Quantum-Enhanced AI (QAI) system combining various quantum algorithms: Quantum Neural Networks (Quantum Neural Network) to recognize patterns, Quantum Supporter Vector Machines (Quantum Supporter Vector Machines) to classify, Quantum Approximate Optimization Algorithm (Quantum Approximate Optimization Algorithm) to schedule production and allocate resources, and Variational Quantum Eigensolver (Variational Quantum Eigensolver) to optimize parameters under noisy settings or conditions into a hybrid quantum-class The framework is a quantum computer based framework that encodes classical manufacturing data in the form of quantum states, using amplitude and qubit-efficient embeddings, allowing representation of high-dimensional features with the parallelism and entanglement of quantum computing. Realworld experimental validation of both turbofan engine degradation and semiconductor manufacturing sensor logs was conducted using IBM Qiskit and PennyLane to ensure reproducibility, with noise-aware training and error mitigation policies applied in simulations on NISQ-era hardware. The proposed QAI framework demonstrated higher predictive accuracy (12 to 15) than state-of-the-art hybrid quantum-classical and classical models, with superior accuracy, memorability, and F1 scores, and showed exceptional capabilities in anomaly detection, failure prediction, and optimisation. The results validate that quantum algorithms can be integrated into a predictive analytics pipeline to overcome classical computation bottlenecks, enhance robustness, and provide actionable insights for making real-time manufacturing decisions. Overall, this paper has demonstrated that quantum-assisted AI can be beneficial in intelligent manufacturing environments, laying the groundwork for expanding practical applications that bridge the gap between the promises of quantum computational power and predictive analytics in the industry.
Anomaly detection, an important task in data analysis, involves identifying observations or events that deviate in some way from the rest of the data. Machine learning techniques have shown success in automating this process by detecting hidden patterns and deviations in large-scale data. The potential of quantum computing for machine learning has been widely recognized, leading to extensive research efforts to develop suitable quantum machine learning (QML) algorithms. In particular, the search for QML algorithms for near-term NISQ devices is in full swing. However, NISQ devices pose additional challenges due to their limited qubit coherence times, low number of qubits, and high error rates. Kernel methods based on quantum kernel estimation have emerged as a promising approach to QML on NISQ devices, offering theoretical guarantees, versatility, and compatibility with NISQ constraints. Especially support vector machines (SVM) utilizing quantum kernel estimation have shown success in various supervised learning tasks. However, in the context of anomaly detection, semisupervised learning is of great relevance, and yet there is limited research published in this area. This paper introduces an approach to semisupervised anomaly detection based on the reconstruction loss of a support vector regression (SVR) with quantum kernel. This novel model is an alternative to the variational quantum and quantum kernel one-class classifiers, and is compared to a quantum autoencoder as quantum baseline and a SVR with radial-basis-function (RBF) kernel as well as a classical autoencoder as classical baselines. The models are benchmarked extensively on 10 real-world anomaly detection data sets and one toy data set and it is shown that our SVR model with quantum kernel performs better than the SVR with RBF kernel as well as all other models, achieving highest mean AUC over all data sets. In addition, our QSVR outperforms the Quantum autoencoder on 9 out of 11 data sets.
Quantum algorithms for predictive analytics is important across various sectors such as finance, healthcare, marketing and supply chain management where accurate predictions can be used to optimise operations and decision-making processes. This chapter addresses Specific quantum machine learning algorithms such as Quantum Support Vector Machines (QSVM), Quantum K-nearest Neighbors (QKNN) and Quantum Neural Networks (QNN). These algorithms enhance classification, regression and complex predictive tasks by accelerating training processes and managing large feature spaces effectively. And, to this, quantum optimization algorithms like Quantum Approximate Optimization Algorithm (QAOA) and Variational Quantum Eigensolver (VQE) are mentioned for their role in optimising model parameters and improving prediction accuracy. In addition, it discusses quantum data encoding methods, such as Quantum Fourier Transform (QFT) and amplitude encoding. Along with this, it acknowledges significant challenges such as hardware limitations, algorithmic complexity and practical integration issues.
Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.