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:2604.03482

Learning high-dimensional quantum entanglement through physics-guided neural networks

Yang Xu, Hao Zhang, Wenwen Zhang, Luchang Niu, Mahtab Amooei, Sergio Carbajo, Robert W. Boyd

quant-phSubmitted Apr 3, 2026• Announced Apr 7, 2026• 0 citations• Revised Jun 9, 2026 (v2)
v1 · Apr 3, 2026v2 · Jun 9, 2026
arXivHTMLPDF

Abstract

High-gain spontaneous parametric down-conversion (SPDC) produces bright squeezed vacuum with rich high-dimensional entanglement, but its output is inherently multimodal and non-perturbative, making the full modal characterization a major computational bottleneck. We propose a physics-guided deep neural network that reconstructs the source's modal fingerprint: the high-dimensional correlation signature across radial and azimuthal indices. We designed a FiLM-modulated convolutional architecture that predicts the joint (m,l) distribution, and training is driven by a hybrid loss that couples data-driven metrics (JSD, KL, MSE, Wasserstein) with a soft orbital-angular-momentum (OAM) conservation term, providing an essential inductive bias toward physically consistent solutions. Across gain regimes, our method achieves high-fidelity reconstruction with average JSD of 1.96e-3, WEMD of 1.54e-3, and KL divergence of 7.85e-3, delivering an approximate 128-fold speedup over full numerical simulation and more than 30% accuracy gains over U-Net baselines. These results demonstrate that physics-guided learning, via a soft OAM-conservation regularizer and physically generated training targets, enables rapid and data-efficient modal characterization. Compared with traditional numerical simulation, our mesh-free method has demonstrated good generalization with limited or contaminated training data and has enabled fast "online" prediction of the quantum dynamics of a high-dimensional entanglement system for real-world experimental implementation.

Related papers

Correlation-Pattern-Based Orbital Angular Momentum Entanglement Measurement Through Neural Networks

Jiaxian Zhao, Zhifeng Liu, Chenghou Tu +2 more

Abstract High-dimensional (HD) entanglement of photonic orbital angular momentum (OAM) is pivotal for advancing quantum communication and information processing, but its characterization remains significant challenges due to the complexity of quantum state tomography and experimental limitations such as low photon counts caused by losses. Here, we propose a pre-trained physics-informed neural network (PTPINN) framework that enables efficient and rapid reconstruction of HD-OAM entangled states under low photon counts. Experimental results show that the fidelity of five-dimensional OAM entanglement reaches F = 0.958 ± 0.010 even with an exposure time as short as 50 ms. This highlights the capability of PTPINN to achieve high-precision quantum state reconstruction with limited photons, owing to its innovative designs, thus overcoming the reliance on high photon counts typical of traditional methods. Our method provides a practical and scalable solution for high-fidelity characterization of HD-OAM entanglement in environments with low photon numbers and high noise, paving the way for robust long-distance quantum information transmission.

Chinese Physics Letters•Feb 24, 2025•0 citations

Towards Sample Efficient Entanglement Classification for 3 and 4 Qubit Systems: A Tailored CNN-BiLSTM Approach

Qian Sun, Yuedong Sun, Yu Hu +3 more

Accurate classification of multipartite entanglement in high-dimensional quantum systems is crucial for advancing quantum communication and information processing. However, conventional methods are resource-intensive, and even many machine-learning-based approaches necessitate large training datasets, creating a significant experimental bottleneck for data acquisition. To address this challenge, we propose a hybrid neural network architecture integrating Convolutional and Bidirectional Long Short-Term Memory networks (CNN-BiLSTM). This design leverages CNNs for local feature extraction and BiLSTMs for sequential dependency modeling, enabling robust feature learning from minimal training data. We investigate two fusion paradigms: Architecture 1 (flattening-based) and Architecture 2 (dimensionality-transforming). When trained on only 100 samples, Architecture 2 maintains classification accuracies exceeding 90% for both 3-qubit and 4-qubit systems, demonstrating rapid loss convergence within tens of epochs. Under full-data conditions (400 000 samples), both architectures achieve accuracies above 99.97%. Comparative benchmarks reveal that our CNN-BiLSTM models, especially Architecture 2, consistently outperform standalone CNNs, BiLSTMs, and MLPs in low-data regimes, albeit with increased training time. These results demonstrates that the tailored CNN-BiLSTM fusion significantly alleviates experimental data acquisition burden, offering a practical pathway toward scalable entanglement verification in complex quantum systems.

quant-pharXiv:2601.22562•Feb 2, 2026•0 citations

Sparsity-Driven Entanglement Detection in High-Dimensional Quantum States

Stav Lotan, Hugo Defienne, Ronen Talmon +1 more

The characterization of high-dimensional quantum entanglement is crucial for advanced quantum computing and quantum information algorithms. Traditional methods require extensive data acquisition and suffer from limited visibility due to experimental noise. Here, we introduce a sparsity-driven framework to enhance the detection and certification of high-dimensional entanglement in spatially entangled photon pairs. By applying $\ell_1$-regularized reconstruction to sample covariance matrices obtained from measurements on photons produced via spontaneous parametric down-conversion (SPDC) measurements, we enhance the visibility of the correlation signal while suppressing noise. We demonstrate, using a position-momentum Einstein-Podolsky-Rosen (EPR) entanglement criterion, that this approach enables certification of an entanglement dimensionality that cannot be achieved without regularization. Our method is scalable, simple to use and compatible with existing quantum-optics platforms, thus paves the way for efficient, real-time analysis of high-dimensional quantum states.

quant-pharXiv:2511.12546•Newton•Nov 18, 2025•0 citations

Data-Driven Quantification of Quantum k-Entanglement via Machine Learning

Jie Guo, Jinchuan Hou, Xiaofei Qi +1 more

k-entanglement, including entanglement relative to full separability and genuinely multipartite entanglement, is a fundamental quantum resource in multipartite quantum systems. Its identification and quantification play essential roles in quantum information processing, quantum simulation, and quantum metrology. However, the practical computation of rigorous k-entanglement measures remains highly challenging due to the need for high-dimensional optimization. In this work, we propose a machine-learning-based surrogate framework for approximating the witness-based k-entanglement measure Ew(k,n). The numerical evaluation of the computationally realized quantity E˜w(k,n)(ρ) is reformulated as a supervised regression problem, where the input is the density matrix ρ and the labels are obtained from finite witness databases. The framework combines multilayer perceptrons (MLPs), convolutional neural networks (CNNs), and light gradient boosting machine (LightGBM) through a stacking ensemble. Numerical experiments are performed for 3- and 4-qubit systems as representative demonstrations of the proposed workflow. The results show that the learned models achieve high predictive accuracy in terms of MAE, MSE, and R2, while providing millisecond-level inference for single-state evaluation. Werner state tests serve as symmetric benchmark checks, and an additional four-qubit noisy circuit-generated state family, obtained from finite-depth circuit preparation followed by local amplitude-damping noise, is used as a structured physical test beyond random density matrices. Compared with the optimization-based evaluation, the trained surrogate model significantly reduces the computational time while maintaining accuracy within the tested system sizes and data distributions. These results show that the proposed framework provides an efficient numerical surrogate for rapid approximation of witness-based k-entanglement measures, while extensions to larger systems and experimental data require further validation.

Entropy•Jul 22, 2026•0 citations

Deep learning as a tool to distinguish between high orbital angular momentum optical modes

Erin M. Knutson, Sanjaya Lohani, Onur Danaci +2 more

The generation of light containing large degrees of orbital angular momentum (OAM) has recently been demon- strated in both the classical and quantum regimes. Since there is no fundamental limit to how many quanta of OAM a single photon can carry, optical states with an arbitrarily high difference in this quantum number may, in principle, be entangled. This opens the door to investigations into high-dimensional entanglement shared between states in superpositions of nonzero OAM. Additionally, making use of non-zero OAM states can allow for a dramatic increase in the amount of information carried by a single photon, thus increasing the information capacity of a communication channel. In practice, however, it is difficult to differentiate between states with high OAM numbers with high precision. Here we investigate the ability of deep neural networks to differentiate between states that contain large values of OAM. We show that such networks may be used to differentiate be- tween nearby OAM states that contain realistic amounts of noise, with OAM values of up to 100. Additionally, we examine how the classification accuracy scales with the signal-to-noise ratio of images that are used to train the network, as well as those being tested. Finally, we demonstrate the simultaneous classification of < 100 OAM states with greater than 70 % accuracy. We intend to verify our system with experimentally-produced classi- cal OAM states, as well as investigate possibilities that would allow this technique to work in the few-photon quantum regime.

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE•Sep 14, 2016•64 citations

Efficient Measurement of Orbital Angular Momentum Entanglement Using Convolutional Neural Network

Jiaxian Zhao, Min Wang, Shuang‐Yin Huang +4 more

Abstract High‐dimensional (HD) entanglement of photonic orbital angular momentum (OAM) offers significant potential for enhancing channel capacity and improving noise resistance in quantum information processing. However, the challenge of achieving simple and rapid measurement has limited its practical applications. In this work, a quantum state tomography (QST) framework is demonstrated that utilizes convolutional neural networks to rapidly reconstruct the density matrix of OAM entanglement from only two coincidence measurements. The experimental results for a 5D OAM entangled state yield a fidelity of 0.973 ± 0.005. This method is also applicable to mixed OAM entangled states and scenarios with incomplete tomographic measurements. These findings represent a significant step toward implementing high‐speed QST for applications involving HD spatial mode quantum state, whether in free space or integrated systems.

Laser & Photonics Review•Jan 17, 2025•6 citations
High-gain spontaneous parametric down-conversion (SPDC) produces bright squeezed vacuum with rich high-dimensional entanglement, but its output is inherently multimodal and non-perturbative, making the full modal characterization a major computational bottleneck. We propose a physics-guided deep neural network that reconstructs the source's modal fingerprint: the high-dimensional correlation signature across radial and azimuthal indices. We designed a FiLM-modulated convolutional architecture that predicts the joint (m,l) distribution, and training is driven by a hybrid loss that couples data-driven metrics (JSD, KL, MSE, Wasserstein) with a soft orbital-angular-momentum (OAM) conservation term, providing an essential inductive bias toward physically consistent solutions. Across gain regimes, our method achieves high-fidelity reconstruction with average JSD of 1.96e-3, WEMD of 1.54e-3, and KL divergence of 7.85e-3, delivering an approximate 128-fold speedup over full numerical simulation and more than 30% accuracy gains over U-Net baselines. These results demonstrate that physics-guided learning, via a soft OAM-conservation regularizer and physically generated training targets, enables rapid and data-efficient modal characterization. Compared with traditional numerical simulation, our mesh-free method has demonstrated good generalization with limited or contaminated training data and has enabled fast "online" prediction of the quantum dynamics of a high-dimensional entanglement system for real-world experimental implementation.
Abstract High-dimensional (HD) entanglement of photonic orbital angular momentum (OAM) is pivotal for advancing quantum communication and information processing, but its characterization remains significant challenges due to the complexity of quantum state tomography and experimental limitations such as low photon counts caused by losses. Here, we propose a pre-trained physics-informed neural network (PTPINN) framework that enables efficient and rapid reconstruction of HD-OAM entangled states under low photon counts. Experimental results show that the fidelity of five-dimensional OAM entanglement reaches F = 0.958 ± 0.010 even with an exposure time as short as 50 ms. This highlights the capability of PTPINN to achieve high-precision quantum state reconstruction with limited photons, owing to its innovative designs, thus overcoming the reliance on high photon counts typical of traditional methods. Our method provides a practical and scalable solution for high-fidelity characterization of HD-OAM entanglement in environments with low photon numbers and high noise, paving the way for robust long-distance quantum information transmission.
Accurate classification of multipartite entanglement in high-dimensional quantum systems is crucial for advancing quantum communication and information processing. However, conventional methods are resource-intensive, and even many machine-learning-based approaches necessitate large training datasets, creating a significant experimental bottleneck for data acquisition. To address this challenge, we propose a hybrid neural network architecture integrating Convolutional and Bidirectional Long Short-Term Memory networks (CNN-BiLSTM). This design leverages CNNs for local feature extraction and BiLSTMs for sequential dependency modeling, enabling robust feature learning from minimal training data. We investigate two fusion paradigms: Architecture 1 (flattening-based) and Architecture 2 (dimensionality-transforming). When trained on only 100 samples, Architecture 2 maintains classification accuracies exceeding 90% for both 3-qubit and 4-qubit systems, demonstrating rapid loss convergence within tens of epochs. Under full-data conditions (400 000 samples), both architectures achieve accuracies above 99.97%. Comparative benchmarks reveal that our CNN-BiLSTM models, especially Architecture 2, consistently outperform standalone CNNs, BiLSTMs, and MLPs in low-data regimes, albeit with increased training time. These results demonstrates that the tailored CNN-BiLSTM fusion significantly alleviates experimental data acquisition burden, offering a practical pathway toward scalable entanglement verification in complex quantum systems.
The characterization of high-dimensional quantum entanglement is crucial for advanced quantum computing and quantum information algorithms. Traditional methods require extensive data acquisition and suffer from limited visibility due to experimental noise. Here, we introduce a sparsity-driven framework to enhance the detection and certification of high-dimensional entanglement in spatially entangled photon pairs. By applying ℓ1\ell_1ℓ1​-regularized reconstruction to sample covariance matrices obtained from measurements on photons produced via spontaneous parametric down-conversion (SPDC) measurements, we enhance the visibility of the correlation signal while suppressing noise. We demonstrate, using a position-momentum Einstein-Podolsky-Rosen (EPR) entanglement criterion, that this approach enables certification of an entanglement dimensionality that cannot be achieved without regularization. Our method is scalable, simple to use and compatible with existing quantum-optics platforms, thus paves the way for efficient, real-time analysis of high-dimensional quantum states.
k-entanglement, including entanglement relative to full separability and genuinely multipartite entanglement, is a fundamental quantum resource in multipartite quantum systems. Its identification and quantification play essential roles in quantum information processing, quantum simulation, and quantum metrology. However, the practical computation of rigorous k-entanglement measures remains highly challenging due to the need for high-dimensional optimization. In this work, we propose a machine-learning-based surrogate framework for approximating the witness-based k-entanglement measure Ew(k,n). The numerical evaluation of the computationally realized quantity E˜w(k,n)(ρ) is reformulated as a supervised regression problem, where the input is the density matrix ρ and the labels are obtained from finite witness databases. The framework combines multilayer perceptrons (MLPs), convolutional neural networks (CNNs), and light gradient boosting machine (LightGBM) through a stacking ensemble. Numerical experiments are performed for 3- and 4-qubit systems as representative demonstrations of the proposed workflow. The results show that the learned models achieve high predictive accuracy in terms of MAE, MSE, and R2, while providing millisecond-level inference for single-state evaluation. Werner state tests serve as symmetric benchmark checks, and an additional four-qubit noisy circuit-generated state family, obtained from finite-depth circuit preparation followed by local amplitude-damping noise, is used as a structured physical test beyond random density matrices. Compared with the optimization-based evaluation, the trained surrogate model significantly reduces the computational time while maintaining accuracy within the tested system sizes and data distributions. These results show that the proposed framework provides an efficient numerical surrogate for rapid approximation of witness-based k-entanglement measures, while extensions to larger systems and experimental data require further validation.
The generation of light containing large degrees of orbital angular momentum (OAM) has recently been demon- strated in both the classical and quantum regimes. Since there is no fundamental limit to how many quanta of OAM a single photon can carry, optical states with an arbitrarily high difference in this quantum number may, in principle, be entangled. This opens the door to investigations into high-dimensional entanglement shared between states in superpositions of nonzero OAM. Additionally, making use of non-zero OAM states can allow for a dramatic increase in the amount of information carried by a single photon, thus increasing the information capacity of a communication channel. In practice, however, it is difficult to differentiate between states with high OAM numbers with high precision. Here we investigate the ability of deep neural networks to differentiate between states that contain large values of OAM. We show that such networks may be used to differentiate be- tween nearby OAM states that contain realistic amounts of noise, with OAM values of up to 100. Additionally, we examine how the classification accuracy scales with the signal-to-noise ratio of images that are used to train the network, as well as those being tested. Finally, we demonstrate the simultaneous classification of < 100 OAM states with greater than 70 % accuracy. We intend to verify our system with experimentally-produced classi- cal OAM states, as well as investigate possibilities that would allow this technique to work in the few-photon quantum regime.
Abstract High‐dimensional (HD) entanglement of photonic orbital angular momentum (OAM) offers significant potential for enhancing channel capacity and improving noise resistance in quantum information processing. However, the challenge of achieving simple and rapid measurement has limited its practical applications. In this work, a quantum state tomography (QST) framework is demonstrated that utilizes convolutional neural networks to rapidly reconstruct the density matrix of OAM entanglement from only two coincidence measurements. The experimental results for a 5D OAM entangled state yield a fidelity of 0.973 ± 0.005. This method is also applicable to mixed OAM entangled states and scenarios with incomplete tomographic measurements. These findings represent a significant step toward implementing high‐speed QST for applications involving HD spatial mode quantum state, whether in free space or integrated systems.