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:2610.01699

Efficiently Optimizing the Quantum Value of Bell Inequalities using Batched Gradient Descent

Xinyu Xu, Ping Zhu, Weikang Li, Pierre Pocreau, Dalu Ding, Dawei Ding

quant-phSubmitted Oct 1, 2026• Announced Oct 2, 2026• 0 citations
arXivHTMLPDF

Abstract

Exploring the world of Bell inequalities requires numerical methods that scale effectively with the number of parties, inputs, and outputs. Such methods can be used to study large-scale Bell inequalities beyond the reach of analytic methods, to design experiments for violating more complicated Bell inequalities, and to enable applications of Bell inequality violation. Such applications include randomness generation and multi-agent coordination with restricted communication. However, current general purpose optimizers such as the see-saw method struggle to handle Bell inequalities with only a dozen inputs and outputs. We introduce an optimizer for the quantum value of a Bell inequality based on batched gradient descent (BGD). More precisely, our BGD optimizer is a differentiable search over feasible quantum strategies in which states and projective measurements are generated from unconstrained variables. The Bell expression is evaluated by a direct tensor contraction without forming the prohibitively large Bell operator. This formulation makes parallel random restarts inexpensive and is thus naturally implemented by a GPU. We evaluate our optimizer against the see-saw method and find a significant speedup for a wide variety of Bell inequality families. Implemented on a GPU, our BGD optimizer can optimize Bell inequalities in our numerical experiments with more than a thousand inputs or outputs within minutes. For these large-scale Bell inequalities, we also efficiently compute the classical value by enumerating possible strategies using a GPU or expressing the optimization as a mixed-integer linear program. Our optimizers can serve as numerical evaluation tools for Bell experiments and multi-agent coordination problems modeled by Bell inequalities with many inputs and outputs, a common feature of real-world scenarios such as high frequency trading and distributed systems.

AI Summary

The paper addresses numerical lower bounds on the quantum value of Bell inequalities when the numbers of parties, inputs, or outputs make standard see-saw optimization expensive. It formulates a differentiable search over normalized pure states and measurement bases generated by exponentiating skew-Hermitian matrices. Direct tensor contraction evaluates the Bell functional without constructing the dense Bell operator, while batched independent restarts support GPU execution. GPU enumeration with a one-party best-response reduction and a mixed-integer linear program provide complementary classical optimizers. Benchmarks compare CPU/GPU implementations with a modified see-saw implementation, generally using ten restarts and a 300-second budget per restart. On inequalities tailored to maximally entangled states, GPU optimization reaches 1,407 outputs with two inputs, versus 23 outputs for see-saw; both endpoints have 10% restart success. GPU optimization also reaches 13-party MABK and Mermin instances, versus eight parties for see-saw. For a jamming game with 1,820 inputs and four outputs, it obtains winning probability $0.16748$, exceeding the exact classical value $0.15797$ by approximately 6%. Analytic optima and, where tractable, NPA upper bounds assess the numerical results. The method returns feasible lower bounds, not global-optimality guarantees. Its default measurements are rank-one projective; stochastic post-processing does not cover arbitrary fixed-dimensional POVMs. State storage and dense game tensors remain exponential in party number. References for unsolved instances come from separate optimization runs, and the paper does not target very-high-precision optimization. Comparisons concern the specified implementations and resource limits.

Little sign of AI-generated textAI use disclosedUnclear novelty
Claude-family and GPT-family models were also used to assist with writing and proofreading the manuscript, as well as checking the code used for the numerical experiments.

Error found: The abstract overstates the demonstrated classical-value computation: the headline large-scale benchmarks use analytic classical values, and the 41-input, 41-output load-balancing benchmark has only a classical lower bound because neither enumeration nor MILP certified optimality.

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

Comments
54 pages, 8 figures

Related papers

Self-guided method to search maximal Bell violations for unknown quantum states

Li-Kai Yang, Geng Chen, Wen-Hao Zhang +5 more

In recent decades, a great variety of research and applications concerning Bell nonlocality have been developed with the advent of quantum information science. Providing that Bell nonlocality can be revealed by the violation of a family of Bell inequalities, finding maximal Bell violation (MBV) for unknown quantum states becomes an important and inevitable task during Bell experiments. In this paper we introduce a self-guided method to find MBVs for unknown states using a stochastic gradient ascent algorithm (SGA), by parametrizing the corresponding Bell operators. For three investigated systems (two qubit, three qubit, and two qutrit), this method can ascertain the MBV of general two-setting inequalities within 100 iterations. Furthermore, we prove SGA is also feasible when facing more complex Bell scenarios, e.g., $d$-setting $d$-outcome Bell inequality. Moreover, compared to other possible methods, SGA exhibits significant superiority in efficiency, robustness, and versatility.

arXiv:1707.05449•Physical Review A•Nov 7, 2017•4 citations

A quantum gradient descent algorithm for optimizing Gaussian process models

Junpeng Hu, Jinglai Li, Lei Zhang +1 more

Gaussian process regression (GPR) is a nonparametric supervised learning method, widely valued for its ability to quantify uncertainty. Despite its advantages and broad applications, classical GPR implementations face significant scalability challenges, as they involve matrix operations with a cubic complexity in relation to the dataset size. This computational challenge is further compounded by the demand of optimizing the Gaussian Process (GP) model over its hyperparameters, rendering the total computational cost prohibitive for data intensive problems. To address this issue, we propose a quantum gradient descent algorithm to optimize the GP model. Taking advantage of recent advances in quantum algorithms for linear algebra, our algorithm achieves exponential speedup in computing the gradients of the log marginal likelihood. The entire gradient descent process is integrated into the quantum circuit. Through runtime analysis and error bounds, we demonstrate that our approach significantly improves the scalability of GPR model optimization, making it computationally feasible for large-scale applications.

arXiv:2503.17780•Mathematical Models and Methods in Applied Sciences•Apr 1, 2026•2 citations

Optimizing and Certifying Multipartite Permutationally Invariant Bell Inequalities

Jin-Fu Chen, Mengyao Hu, Jordi Tura

Multipartite Bell nonlocality provides a device-independent probe of many-body quantum correlations, but its characterization is limited by the rapid growth of the underlying classical and quantum optimization problems. We develop a scalable method for constructing and certifying permutationally invariant Bell inequalities using only one- and two-body correlators. The construction gives families of inequalities with robust quantum violations for general $m$ measurements as the number of parties $N$ becomes large. To improve robustness against noise, we optimize the ratio of the quantum value to the classical bound for these families in the large-$N$ limit. We then certify the resulting quantum violation using semidefinite programming. For the broad class of Bell inequalities studied here, the infinite-$N$ ratios take simple rational values for finite $m$ and converge to $\coth(1)$ as $m\to\infty$. The optimized inequalities efficiently detect many-body Bell nonlocality with collective measurements, with more measurement settings leading to stronger violations.

quant-pharXiv:2607.08462•Jul 10, 2026•1 citation

Gradient-descent methods for fast quantum state tomography

Akshay Gaikwad, Manuel Sebastian Torres, Shahnawaz Ahmed +1 more

Abstract Quantum state tomography (QST) is a widely employed technique for characterizing the state of a quantum system. However, it is plagued by two fundamental challenges: computational and experimental complexity grows exponentially with the number of qubits, rendering experimental implementation and data post-processing arduous even for moderately sized systems. Here, we introduce gradient-descent (GD) algorithms for the post-processing step of QST in discrete- and continuous-variable systems. To ensure physically valid state reconstruction at each iteration step of the algorithm, we use various density-matrix parameterizations: Cholesky decomposition, Stiefel manifold, and projective normalization. These parameterizations have the added benefit of enabling a rank-controlled ansatz, which simplifies reconstruction when there is prior information about the system. We benchmark the performance of our GD-QST techniques against state-of-the-art methods, including constrained convex optimization, conditional generative adversarial networks, and iterative maximum likelihood estimation. Our comparison focuses on time complexity, iteration counts, data requirements, state rank, and robustness against noise. We find that rank-controlled ansatzes in our stochastic mini-batch GD-QST algorithms effectively handle noisy and incomplete data sets, yielding significantly higher reconstruction fidelity than other methods. Simulations achieving full-rank seven-qubit QST in under three minutes on a standard laptop, with 18 GB of RAM and no dedicated GPU, highlight that GD-QST is computationally more efficient and outperforms other techniques in most scenarios, offering a promising avenue for characterizing noisy intermediate-scale quantum devices. Our Python code for GD-QST algorithms is publicly available at github.com/mstorresh/GD-QST .

arXiv:2503.04526•Quantum Science and Technology•Sep 25, 2025•14 citations

Bounding large-scale Bell inequalities

Luke Mortimer

Bell inequalities are an important tool for studying nonlocality, however, they quickly become computationally intractable as the system size grows. We consider a method for finding an upper bound for the quantum violation of such inequalities by combining the Navascu\'es-Pironio-Ac\'{\i}n (NPA) hierarchy, the method of alternating projections, and a standard memory-efficient optimization algorithm. While our method may not give the tightest upper bound possible, it often does so several orders of magnitude faster than state-of-the-art solvers, with minimal memory usage, thus allowing solutions to problems that would otherwise be intractable. We benchmark using the well-studied ${I}_{3322}$ inequality as well as a more general large-scale randomized inequality ${R}_{xx22}$. For randomized inequalities with 130 inputs either side (a first-level moment matrix of size $261\ifmmode\times\else\texttimes\fi{}261$), our method is $\ensuremath{\sim}100\ifmmode\times\else\texttimes\fi{}$ faster than both mosek and SCS while giving a bound only $\ensuremath{\sim}2%$ above the optimum.

arXiv:2412.08532•Physical Review A•May 27, 2025•4 citations

Optimizing quantum violation for multipartite facet Bell inequalities

Jin-Fu Chen, Mengyao Hu, Jordi Tura

Nonlocality shapes quantum correlations, revealed through the violation of Bell inequalities. The intersection of all valid Bell inequalities is the so-called local polytope. In multipartite systems, characterizing the local polytope quickly becomes an intractable task as the system size increases. Optimizing Bell inequalities to maximize the ratio between their quantum value and classical bound is key to understanding multipartite nonlocality. We propose a gradient-based method for this optimization. Numerical results indicate that local maxima of this ratio typically correspond to facet Bell inequalities of the local polytope. This enables an iterative search for tight and robust Bell inequalities. Applied to permutation-invariant scenarios, the method provides tight Bell inequalities with large quantum violations and facilitates experimental certification of Bell correlations without full knowledge of the local polytope. Moreover, analytical results of the maximum ratio are derived in the thermodynamic limit.

quant-pharXiv:2511.07523•Nov 12, 2025•2 citations
Exploring the world of Bell inequalities requires numerical methods that scale effectively with the number of parties, inputs, and outputs. Such methods can be used to study large-scale Bell inequalities beyond the reach of analytic methods, to design experiments for violating more complicated Bell inequalities, and to enable applications of Bell inequality violation. Such applications include randomness generation and multi-agent coordination with restricted communication. However, current general purpose optimizers such as the see-saw method struggle to handle Bell inequalities with only a dozen inputs and outputs. We introduce an optimizer for the quantum value of a Bell inequality based on batched gradient descent (BGD). More precisely, our BGD optimizer is a differentiable search over feasible quantum strategies in which states and projective measurements are generated from unconstrained variables. The Bell expression is evaluated by a direct tensor contraction without forming the prohibitively large Bell operator. This formulation makes parallel random restarts inexpensive and is thus naturally implemented by a GPU. We evaluate our optimizer against the see-saw method and find a significant speedup for a wide variety of Bell inequality families. Implemented on a GPU, our BGD optimizer can optimize Bell inequalities in our numerical experiments with more than a thousand inputs or outputs within minutes. For these large-scale Bell inequalities, we also efficiently compute the classical value by enumerating possible strategies using a GPU or expressing the optimization as a mixed-integer linear program. Our optimizers can serve as numerical evaluation tools for Bell experiments and multi-agent coordination problems modeled by Bell inequalities with many inputs and outputs, a common feature of real-world scenarios such as high frequency trading and distributed systems.
The paper addresses numerical lower bounds on the quantum value of Bell inequalities when the numbers of parties, inputs, or outputs make standard see-saw optimization expensive. It formulates a differentiable search over normalized pure states and measurement bases generated by exponentiating skew-Hermitian matrices. Direct tensor contraction evaluates the Bell functional without constructing the dense Bell operator, while batched independent restarts support GPU execution. GPU enumeration with a one-party best-response reduction and a mixed-integer linear program provide complementary classical optimizers.
Benchmarks compare CPU/GPU implementations with a modified see-saw implementation, generally using ten restarts and a 300-second budget per restart. On inequalities tailored to maximally entangled states, GPU optimization reaches 1,407 outputs with two inputs, versus 23 outputs for see-saw; both endpoints have 10% restart success. GPU optimization also reaches 13-party MABK and Mermin instances, versus eight parties for see-saw. For a jamming game with 1,820 inputs and four outputs, it obtains winning probability 0.167480.167480.16748, exceeding the exact classical value 0.157970.157970.15797 by approximately 6%. Analytic optima and, where tractable, NPA upper bounds assess the numerical results.
The method returns feasible lower bounds, not global-optimality guarantees. Its default measurements are rank-one projective; stochastic post-processing does not cover arbitrary fixed-dimensional POVMs. State storage and dense game tensors remain exponential in party number. References for unsolved instances come from separate optimization runs, and the paper does not target very-high-precision optimization. Comparisons concern the specified implementations and resource limits.
In recent decades, a great variety of research and applications concerning Bell nonlocality have been developed with the advent of quantum information science. Providing that Bell nonlocality can be revealed by the violation of a family of Bell inequalities, finding maximal Bell violation (MBV) for unknown quantum states becomes an important and inevitable task during Bell experiments. In this paper we introduce a self-guided method to find MBVs for unknown states using a stochastic gradient ascent algorithm (SGA), by parametrizing the corresponding Bell operators. For three investigated systems (two qubit, three qubit, and two qutrit), this method can ascertain the MBV of general two-setting inequalities within 100 iterations. Furthermore, we prove SGA is also feasible when facing more complex Bell scenarios, e.g., ddd-setting ddd-outcome Bell inequality. Moreover, compared to other possible methods, SGA exhibits significant superiority in efficiency, robustness, and versatility.
Gaussian process regression (GPR) is a nonparametric supervised learning method, widely valued for its ability to quantify uncertainty. Despite its advantages and broad applications, classical GPR implementations face significant scalability challenges, as they involve matrix operations with a cubic complexity in relation to the dataset size. This computational challenge is further compounded by the demand of optimizing the Gaussian Process (GP) model over its hyperparameters, rendering the total computational cost prohibitive for data intensive problems. To address this issue, we propose a quantum gradient descent algorithm to optimize the GP model. Taking advantage of recent advances in quantum algorithms for linear algebra, our algorithm achieves exponential speedup in computing the gradients of the log marginal likelihood. The entire gradient descent process is integrated into the quantum circuit. Through runtime analysis and error bounds, we demonstrate that our approach significantly improves the scalability of GPR model optimization, making it computationally feasible for large-scale applications.
Multipartite Bell nonlocality provides a device-independent probe of many-body quantum correlations, but its characterization is limited by the rapid growth of the underlying classical and quantum optimization problems. We develop a scalable method for constructing and certifying permutationally invariant Bell inequalities using only one- and two-body correlators. The construction gives families of inequalities with robust quantum violations for general mmm measurements as the number of parties NNN becomes large. To improve robustness against noise, we optimize the ratio of the quantum value to the classical bound for these families in the large-NNN limit. We then certify the resulting quantum violation using semidefinite programming. For the broad class of Bell inequalities studied here, the infinite-NNN ratios take simple rational values for finite mmm and converge to coth⁡(1)\coth(1)coth(1) as m→∞m\to\inftym→∞. The optimized inequalities efficiently detect many-body Bell nonlocality with collective measurements, with more measurement settings leading to stronger violations.
Abstract Quantum state tomography (QST) is a widely employed technique for characterizing the state of a quantum system. However, it is plagued by two fundamental challenges: computational and experimental complexity grows exponentially with the number of qubits, rendering experimental implementation and data post-processing arduous even for moderately sized systems. Here, we introduce gradient-descent (GD) algorithms for the post-processing step of QST in discrete- and continuous-variable systems. To ensure physically valid state reconstruction at each iteration step of the algorithm, we use various density-matrix parameterizations: Cholesky decomposition, Stiefel manifold, and projective normalization. These parameterizations have the added benefit of enabling a rank-controlled ansatz, which simplifies reconstruction when there is prior information about the system. We benchmark the performance of our GD-QST techniques against state-of-the-art methods, including constrained convex optimization, conditional generative adversarial networks, and iterative maximum likelihood estimation. Our comparison focuses on time complexity, iteration counts, data requirements, state rank, and robustness against noise. We find that rank-controlled ansatzes in our stochastic mini-batch GD-QST algorithms effectively handle noisy and incomplete data sets, yielding significantly higher reconstruction fidelity than other methods. Simulations achieving full-rank seven-qubit QST in under three minutes on a standard laptop, with 18 GB of RAM and no dedicated GPU, highlight that GD-QST is computationally more efficient and outperforms other techniques in most scenarios, offering a promising avenue for characterizing noisy intermediate-scale quantum devices. Our Python code for GD-QST algorithms is publicly available at github.com/mstorresh/GD-QST .
Bell inequalities are an important tool for studying nonlocality, however, they quickly become computationally intractable as the system size grows. We consider a method for finding an upper bound for the quantum violation of such inequalities by combining the Navascu\'es-Pironio-Ac\'{\i}n (NPA) hierarchy, the method of alternating projections, and a standard memory-efficient optimization algorithm. While our method may not give the tightest upper bound possible, it often does so several orders of magnitude faster than state-of-the-art solvers, with minimal memory usage, thus allowing solutions to problems that would otherwise be intractable. We benchmark using the well-studied I3322{I}_{3322}I3322​ inequality as well as a more general large-scale randomized inequality Rxx22{R}_{xx22}Rxx22​. For randomized inequalities with 130 inputs either side (a first-level moment matrix of size $261\ifmmode\times\else\texttimes\fi{}261$), our method is $\ensuremath{\sim}100\ifmmode\times\else\texttimes\fi{}$ faster than both mosek and SCS while giving a bound only $\ensuremath{\sim}2%$ above the optimum.
Nonlocality shapes quantum correlations, revealed through the violation of Bell inequalities. The intersection of all valid Bell inequalities is the so-called local polytope. In multipartite systems, characterizing the local polytope quickly becomes an intractable task as the system size increases. Optimizing Bell inequalities to maximize the ratio between their quantum value and classical bound is key to understanding multipartite nonlocality. We propose a gradient-based method for this optimization. Numerical results indicate that local maxima of this ratio typically correspond to facet Bell inequalities of the local polytope. This enables an iterative search for tight and robust Bell inequalities. Applied to permutation-invariant scenarios, the method provides tight Bell inequalities with large quantum violations and facilitates experimental certification of Bell correlations without full knowledge of the local polytope. Moreover, analytical results of the maximum ratio are derived in the thermodynamic limit.