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

International Joint Conference on Neural Networks (IJCNN)

Semi-Supervised Multichannel Anomaly Detection Using QAOA-Driven Selection and Set Covering

Marco Casalbore, Leonardo Lavagna, Antonello Rosato, Francesca Cuomo, Massimo Panella

Published Jun 1, 2026• 0 citations
Publisher

Abstract

Anomaly detection in multivariate time series remains a challenging task due to high dimensionality, temporal dependencies, and complex inter-variable relationships. In this work, we extend a previously proposed hybrid quantum-classical anomaly detection framework to the multivariate setting. The method formulates anomaly detection as a Quadratic Unconstrained Binary Optimization problem solved via the Quantum Approximate Optimization Algorithm, where multichannel information is implicitly encoded in a problem-dependent cost matrix. The proposed method is validated on a real-world dataset of water smart meter measurements, demonstrating robust and competitive detection performance. The goal is to also enable empirical analysis of solution multiplicity and near-degeneracy effects in the quantum optimization landscape, along with their impact on the induced geometric representations of normality.

DOI
10.1109/ijcnn69854.2026.11729523

Related papers

Time Series Anomaly Detection with Quantum Variational Methods and Set Covering

Marco Casalbore, Leonardo Lavagna, Antonello Rosato +1 more

We propose an enhanced hybrid quantum-classical framework for time series anomaly detection. Building on our previous formulation of the problem as a Quadratic Unconstrained Binary Optimization solved via the Quantum Approximate Optimization Algorithm, we extend the methodology with a statistical model selector and a refined set-covering inference scheme that accounts for temporal-value asymmetries. We analyze scalability in terms of qubit resources, assess feasibility on current quantum devices, and validate the approach on a benchmark of over one hundred heterogeneous time series. The results demonstrate interpretable decisions, robust precision, and competitive performance against both shallow classical local detectors and deep global classical architectures, highlighting the potential of the proposed methodology.

IEEE International Conference on Acoustics Speech and Signal Processing•Apr 21, 2026•0 citations

Hybrid Quantum-Classical Framework for Anomaly Detection in Time Series with QUBO formulation and QAOA

Marco Casalbore, Leonardo Lavagna, Antonello Rosato +1 more

In this work, we introduce a hybrid quantum-classical framework to address anomaly detection problems in time series data using an innovative Quadratic Unconstrained Binary Optimization formulation. The proposed approach integrates density-based and statistical methodologies with the Quantum Approximate Optimization Algorithm, providing a versatile tool for anomaly detection. Moreover, the underlying model is transversal to different kinds of anomalies and can be directly applied to diverse applications, ranging from fault detection to novelty discovery. Experimental results prove that this architecture achieves competitive accuracy compared to classical techniques, with the unique advantage of exploring alternative solutions in parallel due to the properties of quantum computing. This capability enables a deeper understanding of patterns in time series data.

International Joint Conference on Neural Networks (IJCNN)•Jun 30, 2025•2 citations

Quantum Autoencoder for Multivariate Time Series Anomaly Detection

Kilian Tscharke, Maximilian Wendlinger, Afrae Ahouzi +4 more

Anomaly Detection (AD) defines the task of identifying observations or events that deviate from typical - or normal - patterns, a critical capability in IT security for recognizing incidents such as system misconfigurations, malware infections, or cyberattacks. In enterprise environments like SAP HANA Cloud systems, this task often involves monitoring high-dimensional, multivariate time series (MTS) derived from telemetry and log data. With the advent of quantum machine learning offering efficient calculations in high-dimensional latent spaces, many avenues open for dealing with such complex data. One approach is the Quantum Autoencoder (QAE), an emerging and promising method with potential for application in both data compression and AD. However, prior applications of QAEs to time series AD have been restricted to univariate data, limiting their relevance for real-world enterprise systems. In this work, we introduce a novel QAE-based framework designed specifically for MTS AD towards enterprise scale. We theoretically develop and experimentally validate the architecture, demonstrating that our QAE achieves performance competitive with neural-network-based autoencoders while requiring fewer trainable parameters. We evaluate our model on datasets that closely reflect SAP system telemetry and show that the proposed QAE is a viable and efficient alternative for semisupervised AD in real-world enterprise settings.

arXiv:2504.17548•IEEE International Conference on Quantum Computing and Engineering (QCE)•Aug 30, 2025•11 citations

Hierarchical Improvement of Quantum Approximate Optimization Algorithm for Object Detection: (Invited Paper)

Junde Li, Mahabubul Alam, Abdullah Ash Saki +1 more

Quantum Approximate Optimization Algorithm (QAOA) provides approximate solution to combinatorial optimization problems. It encodes the cost function using a$p$-level quantum circuit where each level consists a problem Hamiltonian followed by a mixing Hamiltonian. Despite the promises, few real-world applications (besides the pedagogical MaxCut problem) have exploited QAOA. The success of QAOA relies on the classical optimizer, variational parameter setting, and quantum circuit design and compilation. In this study, we implement QAOA and analyze its performance for a broader Quadratic Unconstrained Binary Optimization (QUBO) formulation to solve real-word applications such as, partially occluded object detection problem. Furthermore, we analyze the effects of above influential factors on QAOA performance. We propose a 3-level improvement of hybrid quantum-classical optimization for object detection. We achieve more than 13X execution speedup by choosing L-BFGS-B as classical optimizer at the first level and 5.50X additional speedup by exploiting parameter symmetry and more than 1.23X acceleration using parameter regression at the second level. We empirically show that the circuit will achieve better fidelity by optimally rescheduling gate operations (especially for deeper circuits) at the third level.

International Symposium on Quality Electronic Design (ISQED)•Mar 1, 2020•10 citations

Quantum-Enhanced Anomaly Detection in Subsea Well Sensors: A Comparative Study of Variational Quantum, Hybrid, and Classical Approaches.

Mohamed Ashraf Mostafa

Reliable anomaly detection in subsea infrastructure is critical, as overlooked faults may result in severe environmental damage and substantial financial losses. Although classical machine learning approaches provide effective solutions, they often require significant computational resources and may struggle to capture the complex, high-dimensional correlations present in multivariate sensor data. In this work, we present a comprehensive evaluation of quantum and hybrid quantum–classical architectures using the 3W dataset, a large-scale industrial benchmark comprising 2,184 real and simulated subsea events. We develop an end-to-end pipeline that transforms raw multivariate time-series signals into quantum-compatible feature representations. Four hybrid architectures are systematically compared against six classical baselines in both supervised and unsupervised settings. Our results demonstrate that a compact hybrid model (Hybrid-A) achieves an F1-score of 0.9737 using only 177 trainable parameters, matching the performance of a 10,000-parameter Random Forest and exhibiting substantially higher parameter efficiency than conventional neural networks. In the unsupervised scenario, the proposed Quantum Autoencoder outperforms its classical counterpart, improving F1-score by 5.35% and ROC-AUC by 15.28% while reducing parameter count by 42%. These findings indicate that quantum-enhanced models can achieve competitive performance with significantly reduced model complexity, highlighting their potential for deployment in resource-constrained, safety-critical edge environments. Keywords: Quantum Machine Learning, Industrial Anomaly Detection, Hybrid Quantum-Classical Models, 3W Dataset, Edge Deployment, Parameter Efficiency

•Feb 14, 2026•0 citations

Quantum-Enhanced Stock Portfolio Optimization Using Normalizing Flows and QAOA: Design, Implementation, and Validation

Chirag Solanki, Parthiv Somaiya, Payal Varangaonkar

This paper presents a hybrid quantum-classical portfolio optimization pipeline that integrates a Quantum Normalizing Flow (QNF) scenario generator with a Quantum Approximate Optimization Algorithm (QAOA) binary asset selector. The objective is to improve discrete stock selection under uncertainty by capturing nonlinear return dependencies with QNF and solving a Quadratic Unconstrained Binary Optimization (QUBO) formulation via QAOA. The methodology is validated on ten large-cap U.S. assets and compared against three classical baselines (Equal Weight, Minimum Variance, Maximum Sharpe), using risk-adjusted metrics (annualized return, volatility, Sharpe ratio) and Monte Carlo statistical testing. Key contributions: 1) A modular QNF-QAOA hybrid optimizer for robust asset selection. 2) A rigorously reproducible evaluation pipeline with quantitative performance and statistical significance analysis. 3) Empirical evidence of competitive riskadjusted returns and conservative drawdown behavior.

International Conference on Intelligent and Sustainable Electronics & Computing Technologies (INSECT)•May 1, 2026•0 citations
Anomaly detection in multivariate time series remains a challenging task due to high dimensionality, temporal dependencies, and complex inter-variable relationships. In this work, we extend a previously proposed hybrid quantum-classical anomaly detection framework to the multivariate setting. The method formulates anomaly detection as a Quadratic Unconstrained Binary Optimization problem solved via the Quantum Approximate Optimization Algorithm, where multichannel information is implicitly encoded in a problem-dependent cost matrix. The proposed method is validated on a real-world dataset of water smart meter measurements, demonstrating robust and competitive detection performance. The goal is to also enable empirical analysis of solution multiplicity and near-degeneracy effects in the quantum optimization landscape, along with their impact on the induced geometric representations of normality.
We propose an enhanced hybrid quantum-classical framework for time series anomaly detection. Building on our previous formulation of the problem as a Quadratic Unconstrained Binary Optimization solved via the Quantum Approximate Optimization Algorithm, we extend the methodology with a statistical model selector and a refined set-covering inference scheme that accounts for temporal-value asymmetries. We analyze scalability in terms of qubit resources, assess feasibility on current quantum devices, and validate the approach on a benchmark of over one hundred heterogeneous time series. The results demonstrate interpretable decisions, robust precision, and competitive performance against both shallow classical local detectors and deep global classical architectures, highlighting the potential of the proposed methodology.
In this work, we introduce a hybrid quantum-classical framework to address anomaly detection problems in time series data using an innovative Quadratic Unconstrained Binary Optimization formulation. The proposed approach integrates density-based and statistical methodologies with the Quantum Approximate Optimization Algorithm, providing a versatile tool for anomaly detection. Moreover, the underlying model is transversal to different kinds of anomalies and can be directly applied to diverse applications, ranging from fault detection to novelty discovery. Experimental results prove that this architecture achieves competitive accuracy compared to classical techniques, with the unique advantage of exploring alternative solutions in parallel due to the properties of quantum computing. This capability enables a deeper understanding of patterns in time series data.
Anomaly Detection (AD) defines the task of identifying observations or events that deviate from typical - or normal - patterns, a critical capability in IT security for recognizing incidents such as system misconfigurations, malware infections, or cyberattacks. In enterprise environments like SAP HANA Cloud systems, this task often involves monitoring high-dimensional, multivariate time series (MTS) derived from telemetry and log data. With the advent of quantum machine learning offering efficient calculations in high-dimensional latent spaces, many avenues open for dealing with such complex data. One approach is the Quantum Autoencoder (QAE), an emerging and promising method with potential for application in both data compression and AD. However, prior applications of QAEs to time series AD have been restricted to univariate data, limiting their relevance for real-world enterprise systems. In this work, we introduce a novel QAE-based framework designed specifically for MTS AD towards enterprise scale. We theoretically develop and experimentally validate the architecture, demonstrating that our QAE achieves performance competitive with neural-network-based autoencoders while requiring fewer trainable parameters. We evaluate our model on datasets that closely reflect SAP system telemetry and show that the proposed QAE is a viable and efficient alternative for semisupervised AD in real-world enterprise settings.
Quantum Approximate Optimization Algorithm (QAOA) provides approximate solution to combinatorial optimization problems. It encodes the cost function using appp-level quantum circuit where each level consists a problem Hamiltonian followed by a mixing Hamiltonian. Despite the promises, few real-world applications (besides the pedagogical MaxCut problem) have exploited QAOA. The success of QAOA relies on the classical optimizer, variational parameter setting, and quantum circuit design and compilation. In this study, we implement QAOA and analyze its performance for a broader Quadratic Unconstrained Binary Optimization (QUBO) formulation to solve real-word applications such as, partially occluded object detection problem. Furthermore, we analyze the effects of above influential factors on QAOA performance. We propose a 3-level improvement of hybrid quantum-classical optimization for object detection. We achieve more than 13X execution speedup by choosing L-BFGS-B as classical optimizer at the first level and 5.50X additional speedup by exploiting parameter symmetry and more than 1.23X acceleration using parameter regression at the second level. We empirically show that the circuit will achieve better fidelity by optimally rescheduling gate operations (especially for deeper circuits) at the third level.
Reliable anomaly detection in subsea infrastructure is critical, as overlooked faults may result in severe environmental damage and substantial financial losses. Although classical machine learning approaches provide effective solutions, they often require significant computational resources and may struggle to capture the complex, high-dimensional correlations present in multivariate sensor data. In this work, we present a comprehensive evaluation of quantum and hybrid quantum–classical architectures using the 3W dataset, a large-scale industrial benchmark comprising 2,184 real and simulated subsea events. We develop an end-to-end pipeline that transforms raw multivariate time-series signals into quantum-compatible feature representations. Four hybrid architectures are systematically compared against six classical baselines in both supervised and unsupervised settings. Our results demonstrate that a compact hybrid model (Hybrid-A) achieves an F1-score of 0.9737 using only 177 trainable parameters, matching the performance of a 10,000-parameter Random Forest and exhibiting substantially higher parameter efficiency than conventional neural networks. In the unsupervised scenario, the proposed Quantum Autoencoder outperforms its classical counterpart, improving F1-score by 5.35% and ROC-AUC by 15.28% while reducing parameter count by 42%. These findings indicate that quantum-enhanced models can achieve competitive performance with significantly reduced model complexity, highlighting their potential for deployment in resource-constrained, safety-critical edge environments. Keywords: Quantum Machine Learning, Industrial Anomaly Detection, Hybrid Quantum-Classical Models, 3W Dataset, Edge Deployment, Parameter Efficiency
This paper presents a hybrid quantum-classical portfolio optimization pipeline that integrates a Quantum Normalizing Flow (QNF) scenario generator with a Quantum Approximate Optimization Algorithm (QAOA) binary asset selector. The objective is to improve discrete stock selection under uncertainty by capturing nonlinear return dependencies with QNF and solving a Quadratic Unconstrained Binary Optimization (QUBO) formulation via QAOA. The methodology is validated on ten large-cap U.S. assets and compared against three classical baselines (Equal Weight, Minimum Variance, Maximum Sharpe), using risk-adjusted metrics (annualized return, volatility, Sharpe ratio) and Monte Carlo statistical testing. Key contributions: 1) A modular QNF-QAOA hybrid optimizer for robust asset selection. 2) A rigorously reproducible evaluation pipeline with quantitative performance and statistical significance analysis. 3) Empirical evidence of competitive riskadjusted returns and conservative drawdown behavior.