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 Conference on Artificial Intelligence (CAI)

Quantum Kernels for SVM Entanglement Detection

Ernesto Acosta, Carlos Cano, Daniel Manzano, Guillermo Botella

Published May 8, 2026• 0 citations
Publisher

Abstract

This work presents a quantum machine learning approach for bipartite entanglement detection in 3×3 quantum systems using a support vector machine with quantum kernels. The classifier is trained to distinguish separable from entangled states, which includes both PPT-detectable and PPT-non-detectable instances, based on classical simulations of parameterized quantum feature maps combined with dimensionality reduction via PCA. Our results show that Pauli-based feature map kernels achieve higher classification accuracy than amplitude embedding at the same PCA dimension, reaching up to 78% accuracy for PCA =8, while amplitude embedding requires PCA =32 to attain comparable performance. We also show that increasing the number of Pauli feature map repetitions does not lead to consistent accuracy improvements, despite significantly increasing circuit depth and computational overhead.

DOI
10.1109/cai68641.2026.11536374

Related papers

Entanglement detection with quantum-inspired kernels and SVMs

Ana Martínez-Sabiote, Michalis Skotiniotis, Jara J. Bermejo-Vega +2 more

Abstract This work presents a machine learning approach based on support vector machines (SVMs) for quantum entanglement detection. Particularly, we focus on bipartite systems of dimensions $$3\times 3$$ 3 × 3 , $$4\times 4$$ 4 × 4 , and $$5\times 5$$ 5 × 5 , where the positive partial transpose criterion (PPT) provides only partial characterization. Using SVMs with quantum-inspired kernels, we develop a classification scheme that distinguishes between separable and entangled states, including PPT-detectable entangled states, and entangled states that evade PPT detection. Our method achieves increasing accuracy with system dimension, reaching $$80\%$$ 80 % , $$90\%$$ 90 % , and nearly $$100\%$$ 100 % for $$3\times 3$$ 3 × 3 , $$4\times 4$$ 4 × 4 , and $$5\times 5$$ 5 × 5 systems, respectively. Our results show that principal component analysis significantly enhances performance for small training sets. The study reveals important practical considerations regarding purity biases in the generation of data for this problem and examines the challenges of implementing these techniques on near-term quantum hardware. Our results demonstrate that machine learning can be an effective alternative to entanglement detection of higher-dimensional systems where conventional entanglement detection methods struggle. Our approach provides improved data generation protocols and can be readily implemented in hybrid classical-quantum architectures, overcoming current limitations.

arXiv:2508.17909•The Journal of Supercomputing•Jan 27, 2026•4 citations

Quantum Kernel-Based Classification: An Optimized Quantum Support Vector Classifier Approach

Xiaoman Liu

Quantum machine learning has emerged as a promising frontier, leveraging the principles of quantum computing to enhance classical algorithms. This paper proposes an optimized Quantum Support Vector Classifier (QSVC) for the classification of the MNIST dataset (a benchmark in handwritten digit recognition). We employ a Chebyshev Parameterized Quantum Circuit (PQC) with an increased number of qubits and layers, and utilize an existing projected quantum kernel method to effectively capture the complex patterns in the data. To enhance model robustness, data augmentation techniques are applied, and dimensionality reduction is performed using Principal Component Analysis (PCA). The proposed QSVC demonstrates competitive accuracy on both the training and testing datasets, highlighting the potential of quantum kernel methods in practical machine learning applications. A comparative analysis with classical Support Vector Machines (SVM) demonstrates the feasibility of quantum methods for high-dimensional classification tasks. Additionally, we investigate the impact of varying the number of qubits in the quantum circuit, providing in-depth insights into the scalability and performance trade-offs of quantum kernel methods.

IEEE International Conference on Frontiers Technology of Information and Computer (ICFTIC)•Dec 13, 2024•0 citations

From Theory to Practice: Quantum Kernels in Support Vector Classification for Improved Performance

Nhat-Trinh Le, Nhat Tan Thai, Cao Vu Bui

Quantum computing offers a powerful new paradigm for enhancing classical machine learning, particularly in classification tasks. In this paper, we examine the use of quantum kernels within Support Vector Classification (SVC) and compare their performance to classical SVC. Our experiments focus on the Iris and Wine datasets. For the Wine dataset, we applied Principal Component Analysis (PCA), reducing the feature space from 13 to 4 dimensions to improve computational efficiency. The results demonstrate that Quantum-enhanced SVC (QSVC) outperforms classical SVC, achieving an accuracy improvement from 96.15% with classical SVM to 98.05% with QSVC for the Wine dataset. Similarly, for the Iris dataset, classical SVC achieves near-perfect accuracy at 100%, while QSVC closely follows with 97.25% accuracy. These results highlight the capability of quantum kernels to improve both accuracy and efficiency in classification tasks, providing a promising direction for scalable quantum-classical hybrid systems in machine learning.

RIVF International Conference on Computing and Communication Technologies (RIVF)•Dec 21, 2024•2 citations

Effect of Repetitions and Entanglement on Performance of Pauli Feature Map

Annika Daspal

Quantum Support Vector Classifier (QSVC) is a quantum machine learning model that solves classification problems. QSVC leverages quantum advantage by utilizing a quantum kernel. A quantum kernel requires a Pauli Feature Map (a data encoding circuit) to encode the classical data into the quantum state space. A Pauli Feature Map's performance can be impacted by two of its key attributes-repetitions and entanglement. In this work, we study the performance of Pauli Feature Map-in terms of QSVC execution time and accuracy-for different combinations of repetitions and entanglement on two different datasets.

IEEE International Conference on Quantum Computing and Engineering (QCE)•Sep 17, 2023•0 citations

A novel RX-RZ Entangled Feature Map Kernel for Fraud Detection using Quantum Support Vector Machine Classifier

Jaydeep Rath, Priya Das, Amit Kumar Mishra

This study proposes a quantum-kernel framework based on a Quantum Support Vector Machine (QSVM) with a novel XYZ feature map integrating RY and RZ rotations for encoding, full all-to-all CX entanglement, and an additional RX rotation layer to obtain higher-order information. The performance of the proposed method is evaluated using a publicly available dataset and benchmarked this embedding against standard Z, ZZ, and Pauli feature maps under matched circuit depth and resource budgets. Across settings, the suggested XYZ map yields the highest performance with accuracy of 93.2%, precision of 0.934, recall of 0.925 and G-mean of 0.93. These results indicate that fully entangling quantum embeddings can enhance the approximation capacity of quantum-kernel for real-world anomaly detection and provide a path toward quantum-assisted fraud analytics.

IEEE International Conference on Applied Electromagnetics, Signal Processing, & Communication (AESPC)•Dec 5, 2025•0 citations

Entanglement classification via witness operators generated by support vector machine

Claudio Sanavio, Edoardo Tignone, Elisa Ercolessi

Abstract Although entanglement is a basic resource for reaching quantum advantage in many computation and information protocols, we lack a universal recipe for detecting it, with analytical results obtained for low-dimensional systems and few special cases of higher-dimensional systems. In this work, we use a machine learning algorithm, the support vector machine with polynomial kernel, to classify separable and entangled states. We apply it to two-qubit and three-qubit systems, and we show that, after training, the support vector machine is able to recognize if a random state is entangled with an accuracy up to $$92\%$$ 92 % for the two-qubit system and up to $$98\%$$ 98 % for the three-qubit system. We also describe why and in what regime the support vector machine algorithm is able to implement the evaluation of an entanglement witness operator applied to many copies of the state, and we describe how we can translate this procedure into a quantum circuit.

arXiv:2301.06759•The European Physical Journal Plus•Oct 21, 2023•12 citations
This work presents a quantum machine learning approach for bipartite entanglement detection in 3×3 quantum systems using a support vector machine with quantum kernels. The classifier is trained to distinguish separable from entangled states, which includes both PPT-detectable and PPT-non-detectable instances, based on classical simulations of parameterized quantum feature maps combined with dimensionality reduction via PCA. Our results show that Pauli-based feature map kernels achieve higher classification accuracy than amplitude embedding at the same PCA dimension, reaching up to 78% accuracy for PCA =8, while amplitude embedding requires PCA =32 to attain comparable performance. We also show that increasing the number of Pauli feature map repetitions does not lead to consistent accuracy improvements, despite significantly increasing circuit depth and computational overhead.
Abstract This work presents a machine learning approach based on support vector machines (SVMs) for quantum entanglement detection. Particularly, we focus on bipartite systems of dimensions 3×33\times 33×3 3 × 3 , 4×44\times 44×4 4 × 4 , and 5×55\times 55×5 5 × 5 , where the positive partial transpose criterion (PPT) provides only partial characterization. Using SVMs with quantum-inspired kernels, we develop a classification scheme that distinguishes between separable and entangled states, including PPT-detectable entangled states, and entangled states that evade PPT detection. Our method achieves increasing accuracy with system dimension, reaching 80%80\%80% 80 % , 90%90\%90% 90 % , and nearly 100%100\%100% 100 % for 3×33\times 33×3 3 × 3 , 4×44\times 44×4 4 × 4 , and 5×55\times 55×5 5 × 5 systems, respectively. Our results show that principal component analysis significantly enhances performance for small training sets. The study reveals important practical considerations regarding purity biases in the generation of data for this problem and examines the challenges of implementing these techniques on near-term quantum hardware. Our results demonstrate that machine learning can be an effective alternative to entanglement detection of higher-dimensional systems where conventional entanglement detection methods struggle. Our approach provides improved data generation protocols and can be readily implemented in hybrid classical-quantum architectures, overcoming current limitations.
Quantum machine learning has emerged as a promising frontier, leveraging the principles of quantum computing to enhance classical algorithms. This paper proposes an optimized Quantum Support Vector Classifier (QSVC) for the classification of the MNIST dataset (a benchmark in handwritten digit recognition). We employ a Chebyshev Parameterized Quantum Circuit (PQC) with an increased number of qubits and layers, and utilize an existing projected quantum kernel method to effectively capture the complex patterns in the data. To enhance model robustness, data augmentation techniques are applied, and dimensionality reduction is performed using Principal Component Analysis (PCA). The proposed QSVC demonstrates competitive accuracy on both the training and testing datasets, highlighting the potential of quantum kernel methods in practical machine learning applications. A comparative analysis with classical Support Vector Machines (SVM) demonstrates the feasibility of quantum methods for high-dimensional classification tasks. Additionally, we investigate the impact of varying the number of qubits in the quantum circuit, providing in-depth insights into the scalability and performance trade-offs of quantum kernel methods.
Quantum computing offers a powerful new paradigm for enhancing classical machine learning, particularly in classification tasks. In this paper, we examine the use of quantum kernels within Support Vector Classification (SVC) and compare their performance to classical SVC. Our experiments focus on the Iris and Wine datasets. For the Wine dataset, we applied Principal Component Analysis (PCA), reducing the feature space from 13 to 4 dimensions to improve computational efficiency. The results demonstrate that Quantum-enhanced SVC (QSVC) outperforms classical SVC, achieving an accuracy improvement from 96.15% with classical SVM to 98.05% with QSVC for the Wine dataset. Similarly, for the Iris dataset, classical SVC achieves near-perfect accuracy at 100%, while QSVC closely follows with 97.25% accuracy. These results highlight the capability of quantum kernels to improve both accuracy and efficiency in classification tasks, providing a promising direction for scalable quantum-classical hybrid systems in machine learning.
Quantum Support Vector Classifier (QSVC) is a quantum machine learning model that solves classification problems. QSVC leverages quantum advantage by utilizing a quantum kernel. A quantum kernel requires a Pauli Feature Map (a data encoding circuit) to encode the classical data into the quantum state space. A Pauli Feature Map's performance can be impacted by two of its key attributes-repetitions and entanglement. In this work, we study the performance of Pauli Feature Map-in terms of QSVC execution time and accuracy-for different combinations of repetitions and entanglement on two different datasets.
This study proposes a quantum-kernel framework based on a Quantum Support Vector Machine (QSVM) with a novel XYZ feature map integrating RY and RZ rotations for encoding, full all-to-all CX entanglement, and an additional RX rotation layer to obtain higher-order information. The performance of the proposed method is evaluated using a publicly available dataset and benchmarked this embedding against standard Z, ZZ, and Pauli feature maps under matched circuit depth and resource budgets. Across settings, the suggested XYZ map yields the highest performance with accuracy of 93.2%, precision of 0.934, recall of 0.925 and G-mean of 0.93. These results indicate that fully entangling quantum embeddings can enhance the approximation capacity of quantum-kernel for real-world anomaly detection and provide a path toward quantum-assisted fraud analytics.
Abstract Although entanglement is a basic resource for reaching quantum advantage in many computation and information protocols, we lack a universal recipe for detecting it, with analytical results obtained for low-dimensional systems and few special cases of higher-dimensional systems. In this work, we use a machine learning algorithm, the support vector machine with polynomial kernel, to classify separable and entangled states. We apply it to two-qubit and three-qubit systems, and we show that, after training, the support vector machine is able to recognize if a random state is entangled with an accuracy up to 92%92\%92% 92 % for the two-qubit system and up to 98%98\%98% 98 % for the three-qubit system. We also describe why and in what regime the support vector machine algorithm is able to implement the evaluation of an entanglement witness operator applied to many copies of the state, and we describe how we can translate this procedure into a quantum circuit.