Descubriendo el potencial de la computación cuántica
Conocimiento al servicio de la ciencia y la innovación
Del desarrollo a la aplicación
Encuentra las últimas publicaciones disponibles para cada una de nuestras actividades
Arquitecturas de simulación cuántica en sistemas HPC
IA cuántica para aplicaciones científicas e industriales
IA cuántica para la caracterización del ruido de los ordenadores cuánticos
Algoritmos de aprendizaje automático de inspiración cuántica y sus aplicaciones en la privacidad
Algoritmos de cadenas de Markov cuánticas
Computación cuántica analógico-digital
Algoritmos genéticos cuánticos, variacionales y adiabáticos y sus aplicaciones
Aplicaciones de la IA cuántica a la química cuántica, ciencia de materiales y finanzas
“Quantum reservoir computing” para aprendizaje automático
Algoritmos de corrección cuántica de errores e IA
Modelos de clasificación y generativos cuánticos aplicados a física de partículas y medicina
Análisis de complejidad computacional de algoritmos cuánticos e IA para analizar procesos cuánticos
Optimización con algoritmos de templado cuántico (“quantum annealing”)
Técnicas de IA y quantum machine learning para la identificación de procesos cuánticos
Desarrollo de software de control de simuladores cuánticos para quantum machine learning
Desarrollo y certificación de algoritmos cuánticos en hardware NISQ
Redes neuronales cuánticas y sus aplicaciones al aprendizaje de procesos cuánticos
Algoritmos cuánticos y de inspiración cuántica para problemas matemáticos complejos
Desarrollo de algoritmos y tecnología de control para simuladores cuánticos
Estudio pormenorizado de las arquitecturas cuánticas y sus desafíos de software y hardware
El trayecto de Quantum Spain
Lee los logros más recientes del proyecto
Casas, B.; Mieldzioć, G. R.; Ahmad, S.; Płodzień, M.; Bruzda, W.; Cervera-Lierta, A.; Życzkowski, K.
Quantum Circuits for High-Dimensional Absolutely Maximally Entangled States Artículo de revista
En: Quantum Science and Technology, 2026.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Quantum Circuits for High-Dimensional Absolutely Maximally Entangled States},
author = {Casas, B. and Mieldzioć, G.R. and Ahmad, S. and Płodzień, M. and Bruzda, W. and Cervera-Lierta, A. and Życzkowski, K. },
url = {https://iopscience.iop.org/article/10.1088/2058-9565/ae46d6},
doi = {10.1088/2058-9565/ae46d6},
year = {2026},
date = {2026-02-26},
urldate = {2025-04-07},
journal = {Quantum Science and Technology},
abstract = {Absolutely maximally entangled (AME) states of multipartite quantum systems exhibit maximal entanglement across all possible bipartitions. These states lead to teleportation protocols that surpass standard teleportation schemes, determine quantum error correction codes and can be used to test performance of current term quantum processors. Several AME states can be constructed from graph states using minimal quantum resources. However, there exist other constructions that depart from the stabilizer formalism. In this work, we present explicit quantum circuits to generate exemplary non-stabilizer AME states of four subsystems with four, six, and eight levels each and analyze their capabilities to perform quantum information tasks.
},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
Absolutely maximally entangled (AME) states of multipartite quantum systems exhibit maximal entanglement across all possible bipartitions. These states lead to teleportation protocols that surpass standard teleportation schemes, determine quantum error correction codes and can be used to test performance of current term quantum processors. Several AME states can be constructed from graph states using minimal quantum resources. However, there exist other constructions that depart from the stabilizer formalism. In this work, we present explicit quantum circuits to generate exemplary non-stabilizer AME states of four subsystems with four, six, and eight levels each and analyze their capabilities to perform quantum information tasks.
Krasimirov-Ivanov, T.; Cervera-Lierta, A.; Stornati, P.; Centrone, F.
Hardware-inspired Continuous Variables Quantum Optical Neural Networks Working paper
2026.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {Hardware-inspired Continuous Variables Quantum Optical Neural Networks},
author = {Krasimirov-Ivanov, T. and Cervera-Lierta, A. and Stornati, P. and Centrone, F.},
url = {https://arxiv.org/abs/2512.05204},
doi = {doi.org/10.48550/arXiv.2512.05204},
year = {2026},
date = {2026-01-14},
abstract = {Continuous-variables (CV) quantum optics is a natural formalism for neural networks (NNs) due to its ability to reproduce the information processing of such trainable interconnected systems. In quantum optics, Gaussian operators induce affine mappings on the quadratures of optical modes while non-Gaussian resources (the challenging piece for physical implementation) originate the nonlinear effects, unlocking quantum analogs of an artificial neuron. This work presents a novel experimentally-feasible framework for continuous-variable quantum optical neural networks (QONNs) developed with available photonic components: coherent states as input encoding, a general Gaussian transformation followed by multi-mode photon subtractions as the processing layer, and homodyne detection as outputs readout. The closed-form expressions of such architecture are derived demonstrating the family of adaptive activations and the quantum-optical neurons that emerge from the amount of photon-subtracted modes, proving that the proposed design satisfies the Universal Approximation Theorem within a single layer. To classically simulate the QONN training, the high-performance QuaNNTO library has been developed based on Wick-Isserlis expansion and Bogoliubov transformations, allowing multi-layer exact expectation values of non-Gaussian states without truncating the infinite-dimensional Hilbert space. Experiments on supervised learning and state-preparation tasks show balanced-resource efficiency with strong expressivity and generalization capabilities, illustrating the potential of the architecture for scalable photonic quantum machine learning and for quantum applications such as complex non-Gaussian gate synthesis.
},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
Continuous-variables (CV) quantum optics is a natural formalism for neural networks (NNs) due to its ability to reproduce the information processing of such trainable interconnected systems. In quantum optics, Gaussian operators induce affine mappings on the quadratures of optical modes while non-Gaussian resources (the challenging piece for physical implementation) originate the nonlinear effects, unlocking quantum analogs of an artificial neuron. This work presents a novel experimentally-feasible framework for continuous-variable quantum optical neural networks (QONNs) developed with available photonic components: coherent states as input encoding, a general Gaussian transformation followed by multi-mode photon subtractions as the processing layer, and homodyne detection as outputs readout. The closed-form expressions of such architecture are derived demonstrating the family of adaptive activations and the quantum-optical neurons that emerge from the amount of photon-subtracted modes, proving that the proposed design satisfies the Universal Approximation Theorem within a single layer. To classically simulate the QONN training, the high-performance QuaNNTO library has been developed based on Wick-Isserlis expansion and Bogoliubov transformations, allowing multi-layer exact expectation values of non-Gaussian states without truncating the infinite-dimensional Hilbert space. Experiments on supervised learning and state-preparation tasks show balanced-resource efficiency with strong expressivity and generalization capabilities, illustrating the potential of the architecture for scalable photonic quantum machine learning and for quantum applications such as complex non-Gaussian gate synthesis.
Tejedor, M.; Casas, B.; Conejero, J.; Cervera-Lierta, A.; Badia, R. M.
Orchestrating Quantum-HPC Workflows with Distributed Quantum Circuit Cutting Artículo de revista
En: ACM, 2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Orchestrating Quantum-HPC Workflows with Distributed Quantum Circuit Cutting},
author = {Tejedor, M. and Casas, B. and Conejero, J. and Cervera-Lierta, A. and Badia, R.M.},
url = {https://dl.acm.org/doi/10.1145/3731599.3767547},
doi = {doi.org/10.1145/3731599.3767547},
year = {2025},
date = {2025-11-15},
journal = {ACM},
abstract = {Most quantum computers today are constrained by hardware limitations, particularly the number of available qubits, causing significant challenges for executing large-scale quantum algorithms. Circuit cutting has emerged as a key technique to overcome these limitations by decomposing large quantum circuits into smaller subcircuits that can be executed independently and later reconstructed. In this work, we introduce Qdislib, a distributed and flexible library for quantum circuit cutting, designed to seamlessly integrate with hybrid quantum-classical high-performance computing (HPC) systems. Qdislib employs a graph-based representation of quantum circuits to enable efficient partitioning, manipulation and execution, supporting both wire cutting and gate cutting techniques. The library is compatible with multiple quantum computing libraries, including Qiskit and Qibo, and leverages distributed computing frameworks to execute subcircuits across CPUs, GPUs, and quantum processing units (QPUs) in a fully parallelized manner. We present a proof of concept demonstrating how Qdislib enables the distributed execution of quantum circuits across heterogeneous computing resources, showcasing its potential for scalable quantum-classical workflows.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
Most quantum computers today are constrained by hardware limitations, particularly the number of available qubits, causing significant challenges for executing large-scale quantum algorithms. Circuit cutting has emerged as a key technique to overcome these limitations by decomposing large quantum circuits into smaller subcircuits that can be executed independently and later reconstructed. In this work, we introduce Qdislib, a distributed and flexible library for quantum circuit cutting, designed to seamlessly integrate with hybrid quantum-classical high-performance computing (HPC) systems. Qdislib employs a graph-based representation of quantum circuits to enable efficient partitioning, manipulation and execution, supporting both wire cutting and gate cutting techniques. The library is compatible with multiple quantum computing libraries, including Qiskit and Qibo, and leverages distributed computing frameworks to execute subcircuits across CPUs, GPUs, and quantum processing units (QPUs) in a fully parallelized manner. We present a proof of concept demonstrating how Qdislib enables the distributed execution of quantum circuits across heterogeneous computing resources, showcasing its potential for scalable quantum-classical workflows.
Puljak, E.; Pierini, M.; García-Saez, A.
Tensor network for anomaly detection in the latent space of proton collision events at the LHC Artículo de revista
En: Machine Learning: Science and Technology, vol. 6, iss. 4, no 045001, 2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Tensor network for anomaly detection in the latent space of proton collision events at the LHC},
author = {Puljak, E. and Pierini, M. and García-Saez, A.},
url = {https://iopscience.iop.org/article/10.1088/2632-2153/ae0243/meta},
doi = {10.1088/2632-2153/ae0243},
year = {2025},
date = {2025-10-06},
urldate = {2026-10-06},
journal = {Machine Learning: Science and Technology},
volume = {6},
number = {045001},
issue = {4},
abstract = {The pursuit of discovering new phenomena at the Large Hadron Collider (LHC) requires constant innovation in algorithms and technologies. Tensor networks are mathematical models at the intersection of classical and quantum machine learning, which present a promising and efficient alternative for tackling these challenges. In this study, we propose a tensor network-based strategy for anomaly detection at the LHC and demonstrate its superior performance in identifying new phenomena compared to established quantum methods. Our model is a parameterized matrix product state with an isometric feature map, processing a latent representation of simulated LHC data generated by an autoencoder. Our results highlight the potential of tensor networks to enhance new-physics discovery.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
The pursuit of discovering new phenomena at the Large Hadron Collider (LHC) requires constant innovation in algorithms and technologies. Tensor networks are mathematical models at the intersection of classical and quantum machine learning, which present a promising and efficient alternative for tackling these challenges. In this study, we propose a tensor network-based strategy for anomaly detection at the LHC and demonstrate its superior performance in identifying new phenomena compared to established quantum methods. Our model is a parameterized matrix product state with an isometric feature map, processing a latent representation of simulated LHC data generated by an autoencoder. Our results highlight the potential of tensor networks to enhance new-physics discovery.
Costa, E.; Pérez-Obiol, A.; Menéndez, J.; Rios, A.; García-Sáez, A.; Juliá-Díaz, B.
A quantum annealing protocol to solve the nuclear shell model Artículo de revista
En: SciPost Physics, 2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {A quantum annealing protocol to solve the nuclear shell model},
author = {Costa, E. and Pérez-Obiol, A. and Menéndez, J. and Rios, A. and García-Sáez, A. and Juliá-Díaz, B.},
url = {https://scipost.org/10.21468/SciPostPhys.19.2.062},
doi = {doi: 10.21468/SciPostPhys.19.2.062},
year = {2025},
date = {2025-08-29},
journal = {SciPost Physics},
abstract = {The nuclear shell model accurately describes the structure and dynamics of atomic nuclei. However, the exponential scaling of the basis size with the number of degrees of freedom hampers a direct numerical solution for heavy nuclei. In this work, we present a quantum annealing protocol to obtain nuclear ground states. We propose a tailored driver Hamiltonian that preserves a large gap and validate our approach in a dozen nuclei with basis sizes up to 105 using classical simulations of the annealing evolution. We explore the relation between the spectral gap and the total time of the annealing protocol, assessing its accuracy by comparing the fidelity and energy relative error to classical benchmarks. While the nuclear Hamiltonian is non-local and thus challenging to implement in current setups, the estimated computational cost of our annealing protocol on quantum circuits is polynomial in the single-particle basis size, paving the way to study heavier nuclei.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
The nuclear shell model accurately describes the structure and dynamics of atomic nuclei. However, the exponential scaling of the basis size with the number of degrees of freedom hampers a direct numerical solution for heavy nuclei. In this work, we present a quantum annealing protocol to obtain nuclear ground states. We propose a tailored driver Hamiltonian that preserves a large gap and validate our approach in a dozen nuclei with basis sizes up to 105 using classical simulations of the annealing evolution. We explore the relation between the spectral gap and the total time of the annealing protocol, assessing its accuracy by comparing the fidelity and energy relative error to classical benchmarks. While the nuclear Hamiltonian is non-local and thus challenging to implement in current setups, the estimated computational cost of our annealing protocol on quantum circuits is polynomial in the single-particle basis size, paving the way to study heavier nuclei.
Triadú-Galí, J.; Garcia-Saez, A.; Juliá-Díaz, B.; Pérez-Obiol, A.
Probing dielectric breakdown in Mott insulators through current oscillations Artículo de revista
En: SciPost Physics, 2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Probing dielectric breakdown in Mott insulators through current oscillations},
author = {Triadú-Galí, J. and Garcia-Saez, A. and Juliá-Díaz, B. and Pérez-Obiol, A.},
url = {https://scipost.org/10.21468/SciPostPhys.19.1.010},
doi = {doi: 10.21468/SciPostPhys.19.1.010},
year = {2025},
date = {2025-07-07},
journal = {SciPost Physics},
abstract = {We investigate the dielectric breakdown of mesoscopic Mott insulators, a phenomenon where a strong electric field destabilizes the insulating state, resulting in a transition to a metallic phase. Using the Landau-Zener formalism, which models the excitation of a two-level system, we derive a theoretical expression for the threshold value of the field. To validate our predictions, we present an efficient protocol for estimating the charge gap and threshold field via non-equilibrium current oscillations, overcoming the computational limitations of exact diagonalization. Our simulations demonstrate the accuracy of our theoretical formula for systems with small gaps. Moreover, our findings are directly testable in ultracold atomic experiments with ring geometries and artificial gauge fields, as our method uses measurable quantities and relies on already available technologies. This work aims to bridge the gap between theoretical models and experimentally realizable protocols, providing tools to explore non-equilibrium mesoscopic phenomena in strongly correlated quantum systems.
},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
We investigate the dielectric breakdown of mesoscopic Mott insulators, a phenomenon where a strong electric field destabilizes the insulating state, resulting in a transition to a metallic phase. Using the Landau-Zener formalism, which models the excitation of a two-level system, we derive a theoretical expression for the threshold value of the field. To validate our predictions, we present an efficient protocol for estimating the charge gap and threshold field via non-equilibrium current oscillations, overcoming the computational limitations of exact diagonalization. Our simulations demonstrate the accuracy of our theoretical formula for systems with small gaps. Moreover, our findings are directly testable in ultracold atomic experiments with ring geometries and artificial gauge fields, as our method uses measurable quantities and relies on already available technologies. This work aims to bridge the gap between theoretical models and experimentally realizable protocols, providing tools to explore non-equilibrium mesoscopic phenomena in strongly correlated quantum systems.
Farreras, M.; Cervera-Lierta, A.
Simulation of the 1d XY model on a quantum computer Artículo de revista
En: SciPost Physics Lecture Notes, 2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Simulation of the 1d XY model on a quantum computer},
author = {Farreras, M. and Cervera-Lierta, A.},
url = {https://scipost.org/10.21468/SciPostPhysLectNotes.95},
doi = {10.21468/SciPostPhysLectNotes.95},
year = {2025},
date = {2025-06-11},
urldate = {2024-10-28},
journal = {SciPost Physics Lecture Notes},
abstract = {The field of quantum computing has grown rapidly in recent years, both in terms of theoretical advancements and the practical construction of quantum computers. These computers were initially proposed, among other reasons, to efficiently simulate and understand the complexities of quantum physics.
In this paper, we present a comprehensive scheme for the exact simulation of the one-dimensional (1-D) XY model on a quantum computer. We successfully diagonalize the proposed Hamiltonian, enabling access to the complete energy spectrum. Furthermore, we propose a novel approach for designing a quantum circuit that performs exact time evolution.
Among the applications enabled by this approach, we compute the ground- and excited-state energies of the symmetric XY model for spin chains with n=4 and n=8 spins. In addition, we calculate the expected value of the transverse magnetization in the ground state of the transverse Ising model. Both studies reveal a quantum phase transition from an antiferromagnetic to a paramagnetic state.
Finally, we simulate the time evolution of the all-spins-up state in the transverse Ising model. The scalability and high performance of our algorithm make it an ideal candidate for benchmarking quantum computers while also providing a foundation for simulating other integrable models on quantum computing platforms.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
The field of quantum computing has grown rapidly in recent years, both in terms of theoretical advancements and the practical construction of quantum computers. These computers were initially proposed, among other reasons, to efficiently simulate and understand the complexities of quantum physics.
In this paper, we present a comprehensive scheme for the exact simulation of the one-dimensional (1-D) XY model on a quantum computer. We successfully diagonalize the proposed Hamiltonian, enabling access to the complete energy spectrum. Furthermore, we propose a novel approach for designing a quantum circuit that performs exact time evolution.
Among the applications enabled by this approach, we compute the ground- and excited-state energies of the symmetric XY model for spin chains with n=4 and n=8 spins. In addition, we calculate the expected value of the transverse magnetization in the ground state of the transverse Ising model. Both studies reveal a quantum phase transition from an antiferromagnetic to a paramagnetic state.
Finally, we simulate the time evolution of the all-spins-up state in the transverse Ising model. The scalability and high performance of our algorithm make it an ideal candidate for benchmarking quantum computers while also providing a foundation for simulating other integrable models on quantum computing platforms.
In this paper, we present a comprehensive scheme for the exact simulation of the one-dimensional (1-D) XY model on a quantum computer. We successfully diagonalize the proposed Hamiltonian, enabling access to the complete energy spectrum. Furthermore, we propose a novel approach for designing a quantum circuit that performs exact time evolution.
Among the applications enabled by this approach, we compute the ground- and excited-state energies of the symmetric XY model for spin chains with n=4 and n=8 spins. In addition, we calculate the expected value of the transverse magnetization in the ground state of the transverse Ising model. Both studies reveal a quantum phase transition from an antiferromagnetic to a paramagnetic state.
Finally, we simulate the time evolution of the all-spins-up state in the transverse Ising model. The scalability and high performance of our algorithm make it an ideal candidate for benchmarking quantum computers while also providing a foundation for simulating other integrable models on quantum computing platforms.
Tejedor, M.; Casas, B.; Conejero, J.; Cervera-Lierta, A.; R Badia, M.
Distributed Quantum Circuit Cutting for Hybrid Quantum-Classical High-Performance Computing Working paper
2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {Distributed Quantum Circuit Cutting for Hybrid Quantum-Classical High-Performance Computing},
author = {Tejedor, M. and Casas, B. and Conejero, J. and Cervera-Lierta, A. and Badia, R,M. },
url = {https://arxiv.org/pdf/2505.01184},
doi = {doi.org/10.48550/arXiv.2505.01184},
year = {2025},
date = {2025-05-05},
urldate = {2025-05-05},
abstract = {Most quantum computers today are constrained by hardware limitations, particularly the number of available qubits, causing significant challenges for executing large-scale quantum algorithms. Circuit cutting has emerged as a key technique to overcome these limitations by decomposing large quantum circuits into smaller subcircuits that can be executed independently and later reconstructed. In this work, we introduce Qdislib, a distributed and flexible library for quantum circuit cutting, designed to seamlessly integrate with hybrid quantum-classical high-performance computing (HPC) systems. Qdislib employs a graph-based representation of quantum circuits to enable efficient partitioning, manipulation and execution, supporting both wire cutting and gate cutting techniques. The library is compatible with multiple quantum computing libraries, including Qiskit and Qibo, and leverages distributed computing frameworks to execute subcircuits across CPUs, GPUs, and quantum processing units (QPUs) in a fully parallelized manner. We present a proof of concept demonstrating how Qdislib enables the distributed execution of quantum circuits across heterogeneous computing resources, showcasing its potential for scalable quantum-classical workflows.},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
Most quantum computers today are constrained by hardware limitations, particularly the number of available qubits, causing significant challenges for executing large-scale quantum algorithms. Circuit cutting has emerged as a key technique to overcome these limitations by decomposing large quantum circuits into smaller subcircuits that can be executed independently and later reconstructed. In this work, we introduce Qdislib, a distributed and flexible library for quantum circuit cutting, designed to seamlessly integrate with hybrid quantum-classical high-performance computing (HPC) systems. Qdislib employs a graph-based representation of quantum circuits to enable efficient partitioning, manipulation and execution, supporting both wire cutting and gate cutting techniques. The library is compatible with multiple quantum computing libraries, including Qiskit and Qibo, and leverages distributed computing frameworks to execute subcircuits across CPUs, GPUs, and quantum processing units (QPUs) in a fully parallelized manner. We present a proof of concept demonstrating how Qdislib enables the distributed execution of quantum circuits across heterogeneous computing resources, showcasing its potential for scalable quantum-classical workflows.
Puljak, E.; Sanchez-Ramirez, S.; Masot-Llima, S.; Vallès-Muns, J.; Garcia-Saez, A.; Pierini, M.
tn4ml: Tensor Network Training and Customization for Machine Learning Working paper
2025.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {tn4ml: Tensor Network Training and Customization for Machine Learning},
author = {Puljak, E. and Sanchez-Ramirez, S. and Masot-Llima, S. and Vallès-Muns, J. and Garcia-Saez, A. and Pierini, M.},
url = {https://arxiv.org/abs/2502.13090},
doi = {doi.org/10.48550/arXiv.2502.13090},
year = {2025},
date = {2025-02-18},
abstract = {Tensor Networks have emerged as a prominent alternative to neural networks for addressing Machine Learning challenges in foundational sciences, paving the way for their applications to real-life problems. This paper introduces tn4ml, a novel library designed to seamlessly integrate Tensor Networks into optimization pipelines for Machine Learning tasks. Inspired by existing Machine Learning frameworks, the library offers a user-friendly structure with modules for data embedding, objective function definition, and model training using diverse optimization strategies. We demonstrate its versatility through two examples: supervised learning on tabular data and unsupervised learning on an image dataset. Additionally, we analyze how customizing the parts of the Machine Learning pipeline for Tensor Networks influences performance metrics.
},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
Tensor Networks have emerged as a prominent alternative to neural networks for addressing Machine Learning challenges in foundational sciences, paving the way for their applications to real-life problems. This paper introduces tn4ml, a novel library designed to seamlessly integrate Tensor Networks into optimization pipelines for Machine Learning tasks. Inspired by existing Machine Learning frameworks, the library offers a user-friendly structure with modules for data embedding, objective function definition, and model training using diverse optimization strategies. We demonstrate its versatility through two examples: supervised learning on tabular data and unsupervised learning on an image dataset. Additionally, we analyze how customizing the parts of the Machine Learning pipeline for Tensor Networks influences performance metrics.
Casas, B.; Bonet-Monroig, X.; Pérez-Salinas, A.
The role of data-induced randomness in quantum machine learning classification tasks Working paper
2024.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {The role of data-induced randomness in quantum machine learning classification tasks},
author = {Casas, B. and Bonet-Monroig, X. and Pérez-Salinas, A. },
url = {https://arxiv.org/pdf/2411.19281},
doi = {doi.org/10.48550/arXiv.2411.19281},
year = {2024},
date = {2024-11-28},
urldate = {2025-11-28},
abstract = {Quantum machine learning (QML) has surged as a prominent area of research with the objective to go beyond the capabilities of classical machine learning models. A critical aspect of any learning task is the process of data embedding, which directly impacts model performance. Poorly designed data-embedding strategies can significantly impact the success of a learning task. Despite its importance, rigorous analyses of data-embedding effects are limited, leaving many cases without effective assessment methods. In this work, we introduce a metric for binary classification tasks, the class margin, by merging the concepts of average randomness and classification margin. This metric analytically connects data-induced randomness with classification accuracy for a given data-embedding map. We benchmark a range of data-embedding strategies through class margin, demonstrating that data-induced randomness imposes a limit on classification performance. We expect this work to provide a new approach to evaluate QML models by their data-embedding processes, addressing gaps left by existing analytical tools.},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
Quantum machine learning (QML) has surged as a prominent area of research with the objective to go beyond the capabilities of classical machine learning models. A critical aspect of any learning task is the process of data embedding, which directly impacts model performance. Poorly designed data-embedding strategies can significantly impact the success of a learning task. Despite its importance, rigorous analyses of data-embedding effects are limited, leaving many cases without effective assessment methods. In this work, we introduce a metric for binary classification tasks, the class margin, by merging the concepts of average randomness and classification margin. This metric analytically connects data-induced randomness with classification accuracy for a given data-embedding map. We benchmark a range of data-embedding strategies through class margin, demonstrating that data-induced randomness imposes a limit on classification performance. We expect this work to provide a new approach to evaluate QML models by their data-embedding processes, addressing gaps left by existing analytical tools.
Oudot, E.; Massé, G.; Valcarce, X.; Acín, A.
Realistic Bell tests with homodyne measurements Working paper
2024.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {Realistic Bell tests with homodyne measurements},
author = {Oudot, E. and Massé, G. and Valcarce, X. and Acín, A.},
url = {https://arxiv.org/abs/2402.01530},
doi = {doi.org/10.48550/arXiv.2402.01530 Focus to learn more},
year = {2024},
date = {2024-02-02},
urldate = {2024-02-02},
abstract = {We analyze Bell inequalities violations in photonic experiments for which the measurement apparatuses are restricted to homodyne measurements. Through numerical optimization of the Clauser-Horne-Shimony-Holt inequality over homodyne measurements and binning choices, we demonstrate large violations for states with a bounded number of photons. When considering states defined within qubit local subspaces of two Fock states, such as NOON states, a violation is observed solely within the qubit Fock space spanned by zero and two photons. For more generic states, large violations are obtained. Significant violations are observed even for states containing three photons locally and under realistic values of noise and losses. We propose concrete implementations to achieve such violations, opening new avenues for Bell experiments with homodyne detectors.
},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
We analyze Bell inequalities violations in photonic experiments for which the measurement apparatuses are restricted to homodyne measurements. Through numerical optimization of the Clauser-Horne-Shimony-Holt inequality over homodyne measurements and binning choices, we demonstrate large violations for states with a bounded number of photons. When considering states defined within qubit local subspaces of two Fock states, such as NOON states, a violation is observed solely within the qubit Fock space spanned by zero and two photons. For more generic states, large violations are obtained. Significant violations are observed even for states containing three photons locally and under realistic values of noise and losses. We propose concrete implementations to achieve such violations, opening new avenues for Bell experiments with homodyne detectors.
Pérez-Obiol, A.; Romero, A. M.; Menéndez, J.; Rios, A.; García-Sáez, A.; Juliá-Díaz, B.
Nuclear shell-model simulation in digital quantum computers Artículo de revista
En: Scientific Reports, vol. 13, 2023.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Nuclear shell-model simulation in digital quantum computers},
author = {Pérez-Obiol, A. and Romero, A. M. and Menéndez, J. and Rios, A. and García-Sáez, A. and Juliá-Díaz, B. },
url = {https://www.nature.com/articles/s41598-023-39263-7},
doi = {doi.org/10.1038/s41598-023-39263-7},
year = {2023},
date = {2023-07-29},
urldate = {2023-02-07},
journal = {Scientific Reports},
volume = {13},
abstract = {The nuclear shell model is one of the prime many-body methods to study the structure of atomic nuclei, but it is hampered by an exponential scaling on the basis size as the number of particles increases. We present a shell-model quantum circuit design strategy to find nuclear ground states that circumvents this limitation by exploiting an adaptive variational quantum eigensolver algorithm. Our circuit implementation is in excellent agreement with classical shell-model simulations for a dozen of light and medium-mass nuclei, including neon and calcium isotopes. We quantify the circuit depth, width and number of gates to encode realistic shell-model wavefunctions. Our strategy also addresses explicitly energy measurements and the required number of circuits to perform them. Our simulated circuits approach the benchmark results exponentially with a polynomial scaling in quantum resources for each nucleus and configuration space. Our work paves the way for quantum computing shell-model studies across the nuclear chart.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
The nuclear shell model is one of the prime many-body methods to study the structure of atomic nuclei, but it is hampered by an exponential scaling on the basis size as the number of particles increases. We present a shell-model quantum circuit design strategy to find nuclear ground states that circumvents this limitation by exploiting an adaptive variational quantum eigensolver algorithm. Our circuit implementation is in excellent agreement with classical shell-model simulations for a dozen of light and medium-mass nuclei, including neon and calcium isotopes. We quantify the circuit depth, width and number of gates to encode realistic shell-model wavefunctions. Our strategy also addresses explicitly energy measurements and the required number of circuits to perform them. Our simulated circuits approach the benchmark results exponentially with a polynomial scaling in quantum resources for each nucleus and configuration space. Our work paves the way for quantum computing shell-model studies across the nuclear chart.
Pérez-Obiol, A.; Masot-Llima, S.; Romero, A. M.; Menéndez, J.; Rios, A.; García-Sáez, A.; Juliá-Díaz, B.
Quantum entanglement patterns in the structure of atomic nuclei within the nuclear shell model Artículo de revista
En: The European Physical Journal A, vol. 59, no 240, 2023, ISBN: 1434-601X.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{nokey,
title = {Quantum entanglement patterns in the structure of atomic nuclei within the nuclear shell model},
author = {Pérez-Obiol, A. and Masot-Llima, S. and Romero, A. M. and Menéndez, J. and Rios, A. and García-Sáez, A. and Juliá-Díaz, B. },
url = {https://link.springer.com/article/10.1140/epja/s10050-023-01151-z},
doi = {doi.org/10.1140/epja/s10050-023-01151-z},
isbn = {1434-601X},
year = {2023},
date = {2023-07-11},
urldate = {2023-07-11},
journal = {The European Physical Journal A},
volume = {59},
number = {240},
abstract = {Quantum entanglement offers a unique perspective into the underlying structure of strongly-correlated systems such as atomic nuclei. In this paper, we use quantum information tools to analyze the structure of light and medium-mass berillyum, oxygen, neon and calcium isotopes within the nuclear shell model. We use different entanglement metrics, including single-orbital entanglement, mutual information, and von Neumann entropies for different equipartitions of the shell-model valence space and identify mode-entanglement patterns related to the energy, angular momentum and isospin of the nuclear single-particle orbitals. We observe that the single-orbital entanglement is directly related to the number of valence nucleons and the energy structure of the shell, while the mutual information highlights signatures of proton–proton and neutron–neutron pairing, as well as nuclear deformation. Proton and neutron orbitals are weakly entangled by all measures, and in fact have the lowest von Neumann entropies among all possible equipartitions of the valence space. In contrast, orbitals with opposite angular momentum projection have relatively large entropies, especially in spherical nuclei. This analysis provides a guide for designing more efficient quantum algorithms for the noisy intermediate-scale quantum era.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
Quantum entanglement offers a unique perspective into the underlying structure of strongly-correlated systems such as atomic nuclei. In this paper, we use quantum information tools to analyze the structure of light and medium-mass berillyum, oxygen, neon and calcium isotopes within the nuclear shell model. We use different entanglement metrics, including single-orbital entanglement, mutual information, and von Neumann entropies for different equipartitions of the shell-model valence space and identify mode-entanglement patterns related to the energy, angular momentum and isospin of the nuclear single-particle orbitals. We observe that the single-orbital entanglement is directly related to the number of valence nucleons and the energy structure of the shell, while the mutual information highlights signatures of proton–proton and neutron–neutron pairing, as well as nuclear deformation. Proton and neutron orbitals are weakly entangled by all measures, and in fact have the lowest von Neumann entropies among all possible equipartitions of the valence space. In contrast, orbitals with opposite angular momentum projection have relatively large entropies, especially in spherical nuclei. This analysis provides a guide for designing more efficient quantum algorithms for the noisy intermediate-scale quantum era.
Casas, B.; Cervera-Lierta, A.
Multi-dimensional Fourier series with quantum circuits Artículo de revista
En: Physical Review A, vol. 107, iss. 5, pp. 15, 2023.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@article{,
title = {Multi-dimensional Fourier series with quantum circuits},
author = {Casas, B. and Cervera-Lierta, A.},
url = {https://journals.aps.org/pra/abstract/10.1103/PhysRevA.107.062612
Preprint version: https://arxiv.org/abs/2302.03389
},
doi = {10.1103/PhysRevA.107.062612},
year = {2023},
date = {2023-06-29},
urldate = {2023-06-29},
journal = {Physical Review A},
volume = {107},
issue = {5},
pages = {15},
abstract = {Quantum machine learning is the field that aims to integrate machine learning with quantum computation. In recent years, the field has emerged as an active research area with the potential to bring new insights to classical machine learning problems. One of the challenges in the field is to explore the expressibility of parametrized quantum circuits and their ability to be universal function approximators, as classical neural networks are. Recent works have shown that, with a quantum supervised learning model, we can fit any one-dimensional Fourier series, proving their universality. However, models for multidimensional functions have not been explored in the same level of detail. In this work, we study the expressibility of various types of circuit Ansätze that generate multidimensional Fourier series. We found that, for some Ansätze, the degrees of freedom required for fitting such functions grow faster than the available degrees in the Hilbert space generated by the circuits. For example, single-qudit models have limited power to represent arbitrary multidimensional Fourier series. Despite this, we show that we can enlarge the Hilbert space of the circuit by using more qudits or higher local dimensions to meet the degrees of freedom requirements, thus ensuring the universality of the models.},
keywords = {BSC},
pubstate = {published},
tppubtype = {article}
}
Quantum machine learning is the field that aims to integrate machine learning with quantum computation. In recent years, the field has emerged as an active research area with the potential to bring new insights to classical machine learning problems. One of the challenges in the field is to explore the expressibility of parametrized quantum circuits and their ability to be universal function approximators, as classical neural networks are. Recent works have shown that, with a quantum supervised learning model, we can fit any one-dimensional Fourier series, proving their universality. However, models for multidimensional functions have not been explored in the same level of detail. In this work, we study the expressibility of various types of circuit Ansätze that generate multidimensional Fourier series. We found that, for some Ansätze, the degrees of freedom required for fitting such functions grow faster than the available degrees in the Hilbert space generated by the circuits. For example, single-qudit models have limited power to represent arbitrary multidimensional Fourier series. Despite this, we show that we can enlarge the Hilbert space of the circuit by using more qudits or higher local dimensions to meet the degrees of freedom requirements, thus ensuring the universality of the models.
Dawid, Anna; Arnold, Julian; Requena, Borja; Gresch, Alexander; Płodzień, Marcin; Donatella, Kaelan; Nicoli, Kim; Stornati, Paolo; Koch, Rouven; Büttner, Miriam; Okuła, Robert; Muñoz-Gil, Gorka; Vargas-Hernández, Rodrigo A.; Cervera-Lierta, Alba; Carrasquilla, Juan; Dunjko, Vedran; Gabrié, Marylou; Huembeli, Patrick; van Nieuwenburg, Evert; Vicentini, Filippo; Wang, Lei; Wetzel, Sebastian J.; Carleo, Giuseppe; Greplová, Eliška; Krems, Roman; Marquardt, Florian; Tomza, Michał; Lewenstein, Maciej; Dauphin, Alexandre
Modern applications of machine learning in quantum sciences Working paper
2022.
Resumen | Enlaces | BibTeX | Etiquetas: BSC
@workingpaper{nokey,
title = {Modern applications of machine learning in quantum sciences},
author = {Anna Dawid and Julian Arnold and Borja Requena and Alexander Gresch and Marcin Płodzień and Kaelan Donatella and Kim Nicoli and Paolo Stornati and Rouven Koch and Miriam Büttner and Robert Okuła and Gorka Muñoz-Gil and Rodrigo A. Vargas-Hernández and Alba Cervera-Lierta and Juan Carrasquilla and Vedran Dunjko and Marylou Gabrié and Patrick Huembeli and Evert van Nieuwenburg and Filippo Vicentini and Lei Wang and Sebastian J. Wetzel and Giuseppe Carleo and Eliška Greplová and Roman Krems and Florian Marquardt and Michał Tomza and Maciej Lewenstein and Alexandre Dauphin},
url = {https://arxiv.org/abs/2204.04198},
doi = {10.48550/arXiv.2204.04198},
year = {2022},
date = {2022-04-08},
urldate = {2022-04-08},
journal = {Arxiv},
pages = {268},
abstract = {In these Lecture Notes, we provide a comprehensive introduction to the most recent advances in the application of machine learning methods in quantum sciences. We cover the use of deep learning and kernel methods in supervised, unsupervised, and reinforcement learning algorithms for phase classification, representation of many-body quantum states, quantum feedback control, and quantum circuits optimization. Moreover, we introduce and discuss more specialized topics such as differentiable programming, generative models, statistical approach to machine learning, and quantum machine learning.},
keywords = {BSC},
pubstate = {published},
tppubtype = {workingpaper}
}
In these Lecture Notes, we provide a comprehensive introduction to the most recent advances in the application of machine learning methods in quantum sciences. We cover the use of deep learning and kernel methods in supervised, unsupervised, and reinforcement learning algorithms for phase classification, representation of many-body quantum states, quantum feedback control, and quantum circuits optimization. Moreover, we introduce and discuss more specialized topics such as differentiable programming, generative models, statistical approach to machine learning, and quantum machine learning.
