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

Physica Scripta

Self-revealing poisons: loss-guided forensic detection and quantum unlearning of corrupted training data

Oum Gadani, Kandarp Gajjar, Himani Trivedi, Hiren Patel

Published Aug 11, 2026• 0 citations
Publisher

Abstract

Abstract Quantum Computation has entered the Noisy Intermediate State Quantum era, in which quantum machine learning (QML) models built on parameterized quantum circuits are promptly utilized for real-world classification tasks, optimization and simulations because of quantum inherent properties like superposition and entanglement. Despite having such use cases, QML model’s security under adversarial conditions remains poorly understood. One such threat is data poisoning, in which an attacker corrupts the training set before the model ever sees it. Such an attack is escalated further in quantum settings by QU antum I ndiscriminate D ata Poisoning, which exploits the geometry of the quantum feature space. To address it, this paper proposes a self-revealing defense framework built on a two-stage pipeline. The first stage identifies potential corrupted data points using unsupervised hybrid k means _ gmm threshold derived from a clean reference model, enabling reliable detection without requiring any prior knowledge of the underlying attack. The second stage uses the detected samples to guide a quantum unlearning process that removes the influence of corrupted data and restores model integrity. All experiments were carried out on simulations (no real quantum hardware is utilized). Experimentation on the MNIST-4 dataset using PQC-8 confirms the attack is self-revealing, as poisoning intensifies the detection PR-AUC climbs from 0.956 to over 0.999, while receiver operating characteristic AUC remains ⩾ 0.986 throughout, achieving a robust detection and clear separability between clean and poisoned samples. Among five evaluated methods used in the quantum unlearning phase of the framework, GA, SCRUB, and Continued Fine-Tuning (CF) recover accuracy to within 10% of the clean baseline for poison ratios up to ε ⩽ 0.5 , with CF achieving this at roughly half the computational cost of the gradient-based alternatives. Additional experiments on classical baseline further reveals the proposed framework being paradigm agnostic.

DOI
10.1088/1402-4896/ae9810

Related papers

Adversarial Data Poisoning Attacks on Quantum Machine Learning in the NISQ Era

Satwik Kundu, Swaroop Ghosh

With the growing interest in Quantum Machine Learning (QML) and the increasing availability of quantum computers through cloud providers, addressing the potential security risks associated with QML has become an urgent priority. One key concern in the QML domain is the threat of data poisoning attacks in the current quantum cloud setting. Adversarial access to training data could severely compromise the integrity and availability of QML models. Classical data poisoning techniques require significant knowledge and training to generate poisoned data, and lack noise resilience, making them ineffective for QML models in the Noisy Intermediate Scale Quantum (NISQ) era. In this work, we first propose a simple yet effective technique to measure intra-class encoder state similarity (ESS) by analyzing the outputs of encoding circuits. Leveraging this approach, we introduce a \underline{Qu}antum \underline{I}ndiscriminate \underline{D}ata Poisoning attack, QUID. Through extensive experiments conducted in both noiseless and noisy environments (e.g., IBM\_Brisbane's noise), across various architectures and datasets, QUID achieves up to $92\%$ accuracy degradation in model performance compared to baseline models and up to $75\%$ accuracy degradation compared to random label-flipping. We also tested QUID against state-of-the-art classical defenses, with accuracy degradation still exceeding $50\%$, demonstrating its effectiveness. This work represents the first attempt to reevaluate data poisoning attacks in the context of QML.

arXiv:2411.14412•Nov 21, 2024•9 citations

Q-Detection: A Quantum-Classical Hybrid Poisoning Attack Detection Method

Haoqi He, Xuanni Lin, J. Chen +1 more

Data poisoning attacks pose significant threats to machine learning models by introducing malicious data into the training process, thereby degrading model performance or manipulating predictions. Detecting and sifting out poisoned data is an important method to prevent data poisoning attacks. Limited by classical computation frameworks, upcoming larger-scale and more complex datasets may pose difficulties for detection. We introduce the unique speedup of quantum computing for the first time in the task of detecting data poisoning. We present Q-Detection, a quantum-classical hybrid defense method for detecting poisoning attacks. Q-Detection also introduces the Quantum Weight-Assigning Network, which is optimized using quantum computing devices. Experimental results using multiple quantum simulation libraries show that Q-Detection effectively defends against label manipulation and backdoor attacks. The metrics demonstrate that Q-Detection consistently outperforms the baseline methods and is comparable to the state-of-the-art. Theoretical analysis shows that Q-Detection is expected to achieve more than a 20% speedup using quantum computing power.

arXiv:2507.06262•International Joint Conference on Artificial Intelligence•Sep 1, 2025•3 citations

Q-Poison: Quantum Adversarial Attacks against QML-driven Interference Classification in O-RAN

Yared Abera Ergu, Van-Linh Nguyen, Po‐Ching Lin +1 more

The integration of quantum machine learning (QML) within open radio access network (O-RAN) architectures presents a promising frontier for enhancing interference classification and optimizing various 6G network use cases. We introduce a novel hybrid quantum-classical neural network (HQCNN) design that addresses critical challenges in noisy intermediate-scale quantum (NISQ) environments through an innovative feature preprocessing pipeline. By integrating deep convolutional feature extraction via ResNet-18 with principal component analysis (PCA)-based dimensionality reduction, we develop a robust computational framework that mitigates quantum gate errors and enhances computational stability. However, despite these advancements, HQCNNs remain vulnerable to adversarial attacks. Traditional classical backdoor methods are ineffective in the quantum setting due to NISQ noise and gate synthesis complexities, while existing quantum attack techniques suffer from a lack of selectivity and adaptability, making them easy to detect. To address this, we introduce Q-Poison, a novel stealthy attack that leverages parameterized quantum gates to significantly enhance attack success rates. This attack induces systematic misclassification of trigger inputs while maintaining 54.5% clean data accuracy, compared to 78.5% without attack. Our results demonstrate a substantial 1.44-fold reduction in accuracy, highlighting the need for robust defense mechanisms in quantum-enabled networks.

IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN)•May 26, 2025•6 citations

SoK: Critical Evaluation of Quantum Machine Learning for Adversarial Robustness

Saeefa Rubaiyet Nowmi, Jesus Lopez, Md Mahmudul Alam Imon +2 more

Quantum Machine Learning (QML) integrates quantum computational principles into learning algorithms, offering the potential for improved representational capacity and computational efficiency. Nevertheless, the security and robustness of QML systems remain largely underexplored, particularly under adversarial conditions. We present the first comprehensive systematization of adversarial robustness in QML, integrating conceptual organization with empirical evaluation across black-, gray-, and white-box threat models. We implement five representative attacks across all three threat models: a label-flipping data poisoning attack under blackbox; an encoder-level indiscriminate data poisoning attack and a proxy-model-based clean-label backdoor attack under gray-box; and a circuit-level backdoor attack (QTrojan) and gradient-based evasion attacks (FGSM and PGD) under whitebox. We evaluate the attacks using a Quantum Multilayer Perceptron (QMLP) trained on MNIST and AZ-Class across circuit depths of 2, 5, 10 and 50 layers and two encoding schemes (angle and amplitude). Our extensive evaluations reveal a fundamental accuracy-robustness trade-off. In particular, amplitude encoding yields the highest clean accuracy (92.6 % on MNIST, 67% on AZ-Class); however, it collapses under adversarial perturbations and depolarizing noise, while shallow angle-encoded models remain substantially more stable. In addition, QUID is highly effective under noiseless conditions but is weakened by noise, whereas the proxy-model backdoor persists unless the circuit itself is overwhelmed, highlighting that noise is an asymmetric and unreliable passive defense. Furthermore, the circuit-level backdoor fails in the multi-class setting, indicating a scalability constraint. Finally, QMLP models are more robust than Classical Multi-Layer Perceptron (CMLP) models under label-flipping attacks but are substantially more vulnerable to gradient-based evasion, motivating the need for quantumspecific defenses. We conclude by proposing a threat-aware, noise-resilient design framework for secure and robust QML deployment.

arXiv:2511.14989•IEEE Symposium on Security and Privacy (S&P)•May 18, 2026•5 citations

Adversarial Attacks on Hybrid Quantum-Classical Interference Classifier in Intelligent O-RAN

Van-Linh Nguyen, Yared Abera Ergu

Quantum machine learning (QML) is emerging as a key enabler of next-generation artificial intelligence (AI), offering more compact models and enhanced data processing capabilities. However, the integration of QML into AI-enabled network services can introduce new adversarial vulnerabilities, particularly the interface between classical encoders and quantum variational circuits. In this work, we investigate the susceptibility of QML-assisted signal classifiers to adversarial threats in the open radio access network (O-RAN) platforms. We introduce a new family of adversarial attacks, including a novel hybrid quantum-classical poisoning method (QC-Poison), along with hybrid gradient-based attacks (QC-FGSM and QC-PGD). QC-Poison induces long-term misclassification by injecting subtle adversarial, accumulating perturbations in the classical input space that propagate through the quantum encoder, effectively drifting the model’s decision boundary. Evaluation results show that QC-FGSM perturbs inputs based on the hybrid model’s gradients, reducing accuracy from 95.5% to 55.8%, while QC-PGD shows model’s performance reduction to 16.0% by iteratively corrupting quantum circuit parameters via loss maximization. QC-Poison achieves 23.9% accuracy under tight perturbation constraints without accessing training data or internal quantum parameters. The results expose critical blind spots in existing hybrid QML models that can be extended to AI-based features in the O-RAN core services. The study underscores the need for robust quantum-aware defenses that can mitigate stealthy adversarial attacks in distributed and QML-assisted applications in intelligent RAN.

IEEE Transactions on Network and Service Management•Jan 1, 2026•1 citation

SQU: Sensitivity-Guided Quantum Machine Unlearning for Privacy-Preserving ML

Md Akmol Masud, Zannat Hossain Tamim, MD Musa Kalimullah Ratul +1 more

The Right to Be Forgotten (RTBF), as set out in regulations such as the General Data Protection Regulation (GDPR), mandates mechanisms to remove specific data points from trained machine learning models. Machine unlearning is a concept in deep learning that has received significant attention but remains underexplored in Quantum Machine Learning (QML). This paper presents a new model for adaptive data deletion in Variational Quantum Circuits (VQCs), called Sensitivity-Guided Quantum Unlearning (SQU). SQU combines gradient-based sensitivity analysis with a dynamic allocation of privacy budgets and targeted noise injection to eliminate the impact of high-impact samples, then fine-tunes the model on the remaining data. Experiments on three benchmark datasets, Breast Cancer, MNIST, and Fashion-MNIST, indicate that SQU successfully unlearns specified samples without diminishing the model's utility. On the Breast Cancer dataset, SQU matches the accuracy of the gold-standard retrain-from-scratch method (84.4%) while providing a quantifiable and significant privacy benefit, reducing the Membership Inference Attack (MIA) AUC by 0.67%. MNIST and Fashion-MNIST results also indicate that SQU has high utility (97.9% and 79.2%, respectively), consistent with the retraining results. A comparison with a classical neural network ($200 \times$more parameters) further demonstrates that the VQC's compact parameterization enables an exact, complete sensitivity analysis for unlearning — an advantage that is infeasible at the classical scale. In general, SQU provides a computationally efficient, regulatory-compliant solution for privacy-aware QML. Code is available at https://github.com/masud1901/Sensitivity-Guided-Quantum-Machine-Unlearning.

IEEE International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN)•Apr 16, 2026•0 citations
Abstract Quantum Computation has entered the Noisy Intermediate State Quantum era, in which quantum machine learning (QML) models built on parameterized quantum circuits are promptly utilized for real-world classification tasks, optimization and simulations because of quantum inherent properties like superposition and entanglement. Despite having such use cases, QML model’s security under adversarial conditions remains poorly understood. One such threat is data poisoning, in which an attacker corrupts the training set before the model ever sees it. Such an attack is escalated further in quantum settings by QU antum I ndiscriminate D ata Poisoning, which exploits the geometry of the quantum feature space. To address it, this paper proposes a self-revealing defense framework built on a two-stage pipeline. The first stage identifies potential corrupted data points using unsupervised hybrid k means _ gmm threshold derived from a clean reference model, enabling reliable detection without requiring any prior knowledge of the underlying attack. The second stage uses the detected samples to guide a quantum unlearning process that removes the influence of corrupted data and restores model integrity. All experiments were carried out on simulations (no real quantum hardware is utilized). Experimentation on the MNIST-4 dataset using PQC-8 confirms the attack is self-revealing, as poisoning intensifies the detection PR-AUC climbs from 0.956 to over 0.999, while receiver operating characteristic AUC remains ⩾ 0.986 throughout, achieving a robust detection and clear separability between clean and poisoned samples. Among five evaluated methods used in the quantum unlearning phase of the framework, GA, SCRUB, and Continued Fine-Tuning (CF) recover accuracy to within 10% of the clean baseline for poison ratios up to ε ⩽ 0.5 , with CF achieving this at roughly half the computational cost of the gradient-based alternatives. Additional experiments on classical baseline further reveals the proposed framework being paradigm agnostic.
With the growing interest in Quantum Machine Learning (QML) and the increasing availability of quantum computers through cloud providers, addressing the potential security risks associated with QML has become an urgent priority. One key concern in the QML domain is the threat of data poisoning attacks in the current quantum cloud setting. Adversarial access to training data could severely compromise the integrity and availability of QML models. Classical data poisoning techniques require significant knowledge and training to generate poisoned data, and lack noise resilience, making them ineffective for QML models in the Noisy Intermediate Scale Quantum (NISQ) era. In this work, we first propose a simple yet effective technique to measure intra-class encoder state similarity (ESS) by analyzing the outputs of encoding circuits. Leveraging this approach, we introduce a \underline{Qu}antum \underline{I}ndiscriminate \underline{D}ata Poisoning attack, QUID. Through extensive experiments conducted in both noiseless and noisy environments (e.g., IBM\_Brisbane's noise), across various architectures and datasets, QUID achieves up to 92%92\%92% accuracy degradation in model performance compared to baseline models and up to 75%75\%75% accuracy degradation compared to random label-flipping. We also tested QUID against state-of-the-art classical defenses, with accuracy degradation still exceeding 50%50\%50%, demonstrating its effectiveness. This work represents the first attempt to reevaluate data poisoning attacks in the context of QML.
Data poisoning attacks pose significant threats to machine learning models by introducing malicious data into the training process, thereby degrading model performance or manipulating predictions. Detecting and sifting out poisoned data is an important method to prevent data poisoning attacks. Limited by classical computation frameworks, upcoming larger-scale and more complex datasets may pose difficulties for detection. We introduce the unique speedup of quantum computing for the first time in the task of detecting data poisoning. We present Q-Detection, a quantum-classical hybrid defense method for detecting poisoning attacks. Q-Detection also introduces the Quantum Weight-Assigning Network, which is optimized using quantum computing devices. Experimental results using multiple quantum simulation libraries show that Q-Detection effectively defends against label manipulation and backdoor attacks. The metrics demonstrate that Q-Detection consistently outperforms the baseline methods and is comparable to the state-of-the-art. Theoretical analysis shows that Q-Detection is expected to achieve more than a 20% speedup using quantum computing power.
The integration of quantum machine learning (QML) within open radio access network (O-RAN) architectures presents a promising frontier for enhancing interference classification and optimizing various 6G network use cases. We introduce a novel hybrid quantum-classical neural network (HQCNN) design that addresses critical challenges in noisy intermediate-scale quantum (NISQ) environments through an innovative feature preprocessing pipeline. By integrating deep convolutional feature extraction via ResNet-18 with principal component analysis (PCA)-based dimensionality reduction, we develop a robust computational framework that mitigates quantum gate errors and enhances computational stability. However, despite these advancements, HQCNNs remain vulnerable to adversarial attacks. Traditional classical backdoor methods are ineffective in the quantum setting due to NISQ noise and gate synthesis complexities, while existing quantum attack techniques suffer from a lack of selectivity and adaptability, making them easy to detect. To address this, we introduce Q-Poison, a novel stealthy attack that leverages parameterized quantum gates to significantly enhance attack success rates. This attack induces systematic misclassification of trigger inputs while maintaining 54.5% clean data accuracy, compared to 78.5% without attack. Our results demonstrate a substantial 1.44-fold reduction in accuracy, highlighting the need for robust defense mechanisms in quantum-enabled networks.
Quantum Machine Learning (QML) integrates quantum computational principles into learning algorithms, offering the potential for improved representational capacity and computational efficiency. Nevertheless, the security and robustness of QML systems remain largely underexplored, particularly under adversarial conditions. We present the first comprehensive systematization of adversarial robustness in QML, integrating conceptual organization with empirical evaluation across black-, gray-, and white-box threat models. We implement five representative attacks across all three threat models: a label-flipping data poisoning attack under blackbox; an encoder-level indiscriminate data poisoning attack and a proxy-model-based clean-label backdoor attack under gray-box; and a circuit-level backdoor attack (QTrojan) and gradient-based evasion attacks (FGSM and PGD) under whitebox. We evaluate the attacks using a Quantum Multilayer Perceptron (QMLP) trained on MNIST and AZ-Class across circuit depths of 2, 5, 10 and 50 layers and two encoding schemes (angle and amplitude). Our extensive evaluations reveal a fundamental accuracy-robustness trade-off. In particular, amplitude encoding yields the highest clean accuracy (92.6 % on MNIST, 67% on AZ-Class); however, it collapses under adversarial perturbations and depolarizing noise, while shallow angle-encoded models remain substantially more stable. In addition, QUID is highly effective under noiseless conditions but is weakened by noise, whereas the proxy-model backdoor persists unless the circuit itself is overwhelmed, highlighting that noise is an asymmetric and unreliable passive defense. Furthermore, the circuit-level backdoor fails in the multi-class setting, indicating a scalability constraint. Finally, QMLP models are more robust than Classical Multi-Layer Perceptron (CMLP) models under label-flipping attacks but are substantially more vulnerable to gradient-based evasion, motivating the need for quantumspecific defenses. We conclude by proposing a threat-aware, noise-resilient design framework for secure and robust QML deployment.
Quantum machine learning (QML) is emerging as a key enabler of next-generation artificial intelligence (AI), offering more compact models and enhanced data processing capabilities. However, the integration of QML into AI-enabled network services can introduce new adversarial vulnerabilities, particularly the interface between classical encoders and quantum variational circuits. In this work, we investigate the susceptibility of QML-assisted signal classifiers to adversarial threats in the open radio access network (O-RAN) platforms. We introduce a new family of adversarial attacks, including a novel hybrid quantum-classical poisoning method (QC-Poison), along with hybrid gradient-based attacks (QC-FGSM and QC-PGD). QC-Poison induces long-term misclassification by injecting subtle adversarial, accumulating perturbations in the classical input space that propagate through the quantum encoder, effectively drifting the model’s decision boundary. Evaluation results show that QC-FGSM perturbs inputs based on the hybrid model’s gradients, reducing accuracy from 95.5% to 55.8%, while QC-PGD shows model’s performance reduction to 16.0% by iteratively corrupting quantum circuit parameters via loss maximization. QC-Poison achieves 23.9% accuracy under tight perturbation constraints without accessing training data or internal quantum parameters. The results expose critical blind spots in existing hybrid QML models that can be extended to AI-based features in the O-RAN core services. The study underscores the need for robust quantum-aware defenses that can mitigate stealthy adversarial attacks in distributed and QML-assisted applications in intelligent RAN.
The Right to Be Forgotten (RTBF), as set out in regulations such as the General Data Protection Regulation (GDPR), mandates mechanisms to remove specific data points from trained machine learning models. Machine unlearning is a concept in deep learning that has received significant attention but remains underexplored in Quantum Machine Learning (QML). This paper presents a new model for adaptive data deletion in Variational Quantum Circuits (VQCs), called Sensitivity-Guided Quantum Unlearning (SQU). SQU combines gradient-based sensitivity analysis with a dynamic allocation of privacy budgets and targeted noise injection to eliminate the impact of high-impact samples, then fine-tunes the model on the remaining data. Experiments on three benchmark datasets, Breast Cancer, MNIST, and Fashion-MNIST, indicate that SQU successfully unlearns specified samples without diminishing the model's utility. On the Breast Cancer dataset, SQU matches the accuracy of the gold-standard retrain-from-scratch method (84.4%) while providing a quantifiable and significant privacy benefit, reducing the Membership Inference Attack (MIA) AUC by 0.67%. MNIST and Fashion-MNIST results also indicate that SQU has high utility (97.9% and 79.2%, respectively), consistent with the retraining results. A comparison with a classical neural network (200×200 \times200×more parameters) further demonstrates that the VQC's compact parameterization enables an exact, complete sensitivity analysis for unlearning — an advantage that is infeasible at the classical scale. In general, SQU provides a computationally efficient, regulatory-compliant solution for privacy-aware QML. Code is available at https://github.com/masud1901/Sensitivity-Guided-Quantum-Machine-Unlearning.