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:2604.13877

Scalable Quantum Molecular Generation via GPU-Accelerated Tensor-Network Simulation

Yu-Cheng Xiao, Jen-Yu Chang, Tzu-Ling Kuo, Aninda Astuti, Shu-Chi Wu, Ka-Lok Ng, Yun-Yuan Wang, Yu-Ze Chen, Nan-Yow Chen, Tai-Yu Li

quant-phSubmitted Apr 15, 2026• Announced Apr 16, 2026• 0 citations
arXivHTMLPDF

Abstract

We propose Scalable Quantum Molecular Generation (SQMG), a variational quantum-circuit for sampling molecular graphs using chemical priors on atoms and bonds. SQMG assigns a fixed 3-qubit register to each heavy atom and reuses a single 2-qubit bond register to generate bonds sequentially, yielding an ''atom no-reuse, bond reuse'' architecture with linear qubit scaling. Measurement results are mapped to molecular graphs via lightweight classical decoding with structural constraints. In CUDA-Q, we benchmark the state-vector simulation (CPU/GPU) and the tensor-network simulation (GPU). At $N=8$ heavy atoms, the state-vector simulator (GPU) and the tensor-network simulator (GPU) achieve speeds of up to $4.5\times 10^{4}$ and $2.2\times 10^{3}$ over the state-vector (CPU) baseline, respectively. Crucially, tensor-network simulation extends exact simulation to $N=40$ heavy atoms, where state-vector methods become memory-limited. For training, Bayesian optimization outperforms COBYLA on a Validity$\times$Uniqueness objective, and the same architecture supports \textit{de novo} generation, scaffold decoration, and linker design. Overall, SQMG provides a scalable, reproducible testbed for evaluating accelerated tensor-network simulation and future quantum molecular generation algorithms.

Related papers

TREV: Python Library for Efficient Implementations of Variational Quantum Algorithms for Optimization using Tensor Networks

Keun Jun Park, Dheeraj Peddireddy, Vaneet Aggarwal

Variational Quantum Algorithms (VQAs) are a central class of hybrid quantum–classical methods for optimization on noisy intermediate-scale quantum (NISQ) hardware. Their scalability is constrained by the cost of gradient evaluation via the parameter-shift rule, whose complexity grows with both the number of circuit parameters and Pauli terms in the Hamiltonian, quickly dominating runtime for deep circuits or large qubit counts. We present TREV (Tensor-Ring Evaluated Variational algorithms), a Python library that accelerates VQA simulation through batched parameter evaluation within tensor-ring state representations . By amortizing contraction and sampling costs across multiple parameter shifts, TREV exploits GPU-parallel tensor contractions to achieve substantial runtime reduction with tunable memory–throughput trade-offs. Implemented in PyTorch, the framework supports both deterministic contraction-based and stochastic sampling-based evaluations. Benchmarks on Max-Cut, Traveling Salesperson Problem (TSP), and molecular ground-state energy estimation (H 2 , H 4 , LiH, BeH 2 ) show comparable accuracy and up to 97% runtime reduction over sequential evaluation, while outperforming PennyLane, TensorCircuit, and Qiskit in runtime scalability. The chemistry benchmarks further demonstrate that TREV handles arbitrary Pauli Hamiltonians beyond QUBO formulations, broadening its applicability to quantum chemistry.

ACM Transactions on Quantum Computing•Jun 22, 2026•0 citations

TREV: Python Library For Efficient Implementations Of Variational Quantum Algorithms For Optimization Using Tensor Networks

Keun Jun Park

Variational Quantum Algorithms (VQAs) are a central class of hybrid quantum-classical methods for optimization on noisy intermediate-scale quantum (NISQ) hardware. Their scalability is constrained by the cost of gradient evaluation via the parameter-shift rule, whose complexity grows with both the number of circuit parameters and Pauli terms in the Hamiltonian, quickly dominating runtime for deep circuits or large qubit counts. We present TREV (Tensor-Ring Evaluated Variational algorithms), a modular Python library that accelerates VQA simulation through batched parameter evaluation within tensor-ring state representations. By amortizing contraction and sampling costs across multiple parameter shifts, TREV exploits GPU-parallel tensor contractions to achieve substantial runtime reduction with tunable memory-throughput trade-offs. Implemented in PyTorch, the framework supports both deterministic contraction-based and stochastic sampling-based evaluations. Benchmarks on Max-Cut, Traveling Salesperson Problem (TSP), and molecular ground-state energy estimation (H2, H4, LiH, BeH2) show comparable accuracy and up to 97% runtime reduction over sequential evaluation, while outperforming PennyLane, TensorCircuit, and Qiskit in runtime scalability. The chemistry benchmarks further demonstrate that TREV handles arbitrary Pauli Hamiltonians beyond QUBO formulations, broadening its applicability to quantum chemistry.

•Apr 28, 2026•0 citations

Meta-Learning for GPU-Accelerated Quantum Many-Body Problems

Yun-Hsuan Chen, Jen-Yu Chang, Tsung-Wei Huang +1 more

We explore the industrial and scientific applicability of the VQE-LSTM framework by integrating meta-learning with GPU accelerated quantum simulation using NVIDIA's CUDA-Q (CUDAQ) platform. This work demonstrates how an LSTM-FC meta-initialization module can extend the practical reach of the Variational Quantum Eigensolver (VQE) in both chemistry and physics domains. In the chemical regime, the framework predicts ground-state energies of molecular Hamiltonians derived from PySCF, achieving near FCI accuracy while maintaining favorable O(N^2) scaling with molecular size. In the physical counterpart, we applied the same model to quantized Simple Harmonic Motion systems (SHM), successfully reproducing its ground and excited states through VQE and Variational Quantum Deflation (VQD) methods. Benchmark results on NVIDIA GPUs reveal significant speedups over CPU-based implementations, validating CUDAQ's capability to handle large-scale variational workloads efficiently. Overall, this study establishes VQE-LSTM as a viable and scalable approach for GPU accelerated quantum simulation, bridging quantum chemistry and condensed-matter physics through a unified, meta-learned initialization strategy.

quant-pharXiv:2602.15706•Feb 18, 2026•0 citations

Rank-Refined Quantum-Behaved Particle Swarm Optimization for Quantum Molecular Generation

Sing-Yun Wu, Sheng Yun Wu, I-Min Chiang +1 more

This work proposes Rank-Refined Quantum-Behaved Particle Swarm Optimization (RR-QPSO) for high-dimensional parameter search in Quantum Molecular Generation (QMG). RR-QPSO targets the optimization bottleneck caused by expensive objective evaluations, where each candidate parameter vector requires stochastic circuit sampling, bitstring decoding, and molecular evaluation. The method provides a population-based alternative to Bayesian optimization (BO), combining Sobol-based initialization, a rank-refined mean-best update, and fitness-guided refinement based on validity and uniqueness. Experiments use the 9-heavy-atom QMG benchmark with a 134-parameter, 20-qubit CUDA-Q circuit and particle evaluations parallelized across 8 NVIDIA V100 GPUs. With M=64 particles and T=150 iterations, RR-QPSO reaches VxU = 0.930; increasing the swarm size to M=128 further improves the product to 0.942, compared with 0.902 for BO under the same protocol. A multi-objective extension targeting HBA=4 and HBD=3 further shows that RR-QPSO can guide molecular properties while preserving a higher validity--uniqueness product than BO. These results suggest that optimizer-level design can improve QMG without modifying the chemistry-inspired circuit or molecular decoding pipeline.

quant-pharXiv:2607.10284•Jul 14, 2026•0 citations

Tensor network surrogate models for variational quantum computation

Ryo Watanabe, Dries Sels, Joseph Tindall

We adopt a two-dimensional tensor-network (TN) ansatz to simulate variational quantum algorithms on two-dimensional qubit architectures, demonstrating its capability to accurately simulate deep circuits through the Quantum Approximate Optimization Algorithm (QAOA) applied to Ising spin-glass problems on heavy-hexagonal and square lattices. For heavy-hexagonal problems with up to three-body interactions, parameters trained on small instances and transferred to systems an order of magnitude larger improve the sampled energy distribution only up to intermediate depths, indicating a fundamental limit of parameter concentration as a transfer strategy. By extending the training itself with TN simulations on larger system sizes, we avoid local minima and obtain lower-energy samples. Analyses of entanglement growth and importance sampling show that the simulation remains classically feasible with moderate bond dimension. We find that parameter concentration also persists on square lattices, albeit at substantially higher computational cost to perform reliable sampling. Overall, our TN framework not only provides an efficient and controlled framework for benchmarking variational quantum algorithms on two-dimensional lattices, but also serves as an effective surrogate model for training variational algorithms.

quant-pharXiv:2604.20180•Apr 23, 2026•0 citations

Efficient Quantum Circuit Simulation by Tensor Network Methods on Modern GPUs

Feng Ping Pan, Hanfeng Gu, Lülin Kuang +2 more

Efficient simulation of quantum circuits has become indispensable with the rapid development of quantum hardware. The primary simulation methods are based on state vectors and tensor networks. As the number of qubits and quantum gates grows larger in current quantum devices, traditional state-vector-based quantum circuit simulation methods prove inadequate due to the overwhelming size of the Hilbert space and extensive entanglement. Consequently, brutal force tensor network simulation algorithms become the only viable solution in such scenarios. The two main challenges faced in tensor network simulation algorithms are optimal contraction path finding and efficient execution on modern computing devices, with the latter determining the actual efficiency. In this study, we investigate the optimization of such tensor network simulations on modern GPUs and propose general optimization strategies from two aspects: computational efficiency and accuracy. First, we propose to transform critical Einstein summation operations into GEMM operations, leveraging the specific features of tensor network simulations to amplify the efficiency of GPUs. Second, by analyzing the data characteristics of quantum circuits, we employ extended precision to ensure the accuracy of simulation results and mixed precision to fully exploit the potential of GPUs, resulting in faster and more precise simulations. Our numerical experiments demonstrate that our approach can achieve a 3.96× reduction in verification time for random quantum circuit samples in the 18-cycle case of Sycamore, with sustained performance exceeding 21 TFLOPS on one A100. This method can be easily extended to the 20-cycle case, maintaining the same performance, accelerating by 12.5× compared to the state-of-the-art CPU-based results and 4.48–6.78× compared to the state-of-the-art GPU-based results reported in the literature. Furthermore, the strategies presented in this article hold general applicability for accelerating problems that can be tackled by contracting tensor networks such as graphical models and combinatorial optimizations.

arXiv:2310.03978•ACM Transactions on Quantum Computing•Sep 23, 2024•10 citations
We propose Scalable Quantum Molecular Generation (SQMG), a variational quantum-circuit for sampling molecular graphs using chemical priors on atoms and bonds. SQMG assigns a fixed 3-qubit register to each heavy atom and reuses a single 2-qubit bond register to generate bonds sequentially, yielding an ''atom no-reuse, bond reuse'' architecture with linear qubit scaling. Measurement results are mapped to molecular graphs via lightweight classical decoding with structural constraints. In CUDA-Q, we benchmark the state-vector simulation (CPU/GPU) and the tensor-network simulation (GPU). At N=8N=8N=8 heavy atoms, the state-vector simulator (GPU) and the tensor-network simulator (GPU) achieve speeds of up to 4.5×1044.5\times 10^{4}4.5×104 and 2.2×1032.2\times 10^{3}2.2×103 over the state-vector (CPU) baseline, respectively. Crucially, tensor-network simulation extends exact simulation to N=40N=40N=40 heavy atoms, where state-vector methods become memory-limited. For training, Bayesian optimization outperforms COBYLA on a Validity×\times×Uniqueness objective, and the same architecture supports \textit{de novo} generation, scaffold decoration, and linker design. Overall, SQMG provides a scalable, reproducible testbed for evaluating accelerated tensor-network simulation and future quantum molecular generation algorithms.
Variational Quantum Algorithms (VQAs) are a central class of hybrid quantum–classical methods for optimization on noisy intermediate-scale quantum (NISQ) hardware. Their scalability is constrained by the cost of gradient evaluation via the parameter-shift rule, whose complexity grows with both the number of circuit parameters and Pauli terms in the Hamiltonian, quickly dominating runtime for deep circuits or large qubit counts. We present TREV (Tensor-Ring Evaluated Variational algorithms), a Python library that accelerates VQA simulation through batched parameter evaluation within tensor-ring state representations . By amortizing contraction and sampling costs across multiple parameter shifts, TREV exploits GPU-parallel tensor contractions to achieve substantial runtime reduction with tunable memory–throughput trade-offs. Implemented in PyTorch, the framework supports both deterministic contraction-based and stochastic sampling-based evaluations. Benchmarks on Max-Cut, Traveling Salesperson Problem (TSP), and molecular ground-state energy estimation (H 2 , H 4 , LiH, BeH 2 ) show comparable accuracy and up to 97% runtime reduction over sequential evaluation, while outperforming PennyLane, TensorCircuit, and Qiskit in runtime scalability. The chemistry benchmarks further demonstrate that TREV handles arbitrary Pauli Hamiltonians beyond QUBO formulations, broadening its applicability to quantum chemistry.
Variational Quantum Algorithms (VQAs) are a central class of hybrid quantum-classical methods for optimization on noisy intermediate-scale quantum (NISQ) hardware. Their scalability is constrained by the cost of gradient evaluation via the parameter-shift rule, whose complexity grows with both the number of circuit parameters and Pauli terms in the Hamiltonian, quickly dominating runtime for deep circuits or large qubit counts. We present TREV (Tensor-Ring Evaluated Variational algorithms), a modular Python library that accelerates VQA simulation through batched parameter evaluation within tensor-ring state representations. By amortizing contraction and sampling costs across multiple parameter shifts, TREV exploits GPU-parallel tensor contractions to achieve substantial runtime reduction with tunable memory-throughput trade-offs. Implemented in PyTorch, the framework supports both deterministic contraction-based and stochastic sampling-based evaluations. Benchmarks on Max-Cut, Traveling Salesperson Problem (TSP), and molecular ground-state energy estimation (H2, H4, LiH, BeH2) show comparable accuracy and up to 97% runtime reduction over sequential evaluation, while outperforming PennyLane, TensorCircuit, and Qiskit in runtime scalability. The chemistry benchmarks further demonstrate that TREV handles arbitrary Pauli Hamiltonians beyond QUBO formulations, broadening its applicability to quantum chemistry.
We explore the industrial and scientific applicability of the VQE-LSTM framework by integrating meta-learning with GPU accelerated quantum simulation using NVIDIA's CUDA-Q (CUDAQ) platform. This work demonstrates how an LSTM-FC meta-initialization module can extend the practical reach of the Variational Quantum Eigensolver (VQE) in both chemistry and physics domains. In the chemical regime, the framework predicts ground-state energies of molecular Hamiltonians derived from PySCF, achieving near FCI accuracy while maintaining favorable O(N^2) scaling with molecular size. In the physical counterpart, we applied the same model to quantized Simple Harmonic Motion systems (SHM), successfully reproducing its ground and excited states through VQE and Variational Quantum Deflation (VQD) methods. Benchmark results on NVIDIA GPUs reveal significant speedups over CPU-based implementations, validating CUDAQ's capability to handle large-scale variational workloads efficiently. Overall, this study establishes VQE-LSTM as a viable and scalable approach for GPU accelerated quantum simulation, bridging quantum chemistry and condensed-matter physics through a unified, meta-learned initialization strategy.
This work proposes Rank-Refined Quantum-Behaved Particle Swarm Optimization (RR-QPSO) for high-dimensional parameter search in Quantum Molecular Generation (QMG). RR-QPSO targets the optimization bottleneck caused by expensive objective evaluations, where each candidate parameter vector requires stochastic circuit sampling, bitstring decoding, and molecular evaluation. The method provides a population-based alternative to Bayesian optimization (BO), combining Sobol-based initialization, a rank-refined mean-best update, and fitness-guided refinement based on validity and uniqueness. Experiments use the 9-heavy-atom QMG benchmark with a 134-parameter, 20-qubit CUDA-Q circuit and particle evaluations parallelized across 8 NVIDIA V100 GPUs. With M=64 particles and T=150 iterations, RR-QPSO reaches VxU = 0.930; increasing the swarm size to M=128 further improves the product to 0.942, compared with 0.902 for BO under the same protocol. A multi-objective extension targeting HBA=4 and HBD=3 further shows that RR-QPSO can guide molecular properties while preserving a higher validity--uniqueness product than BO. These results suggest that optimizer-level design can improve QMG without modifying the chemistry-inspired circuit or molecular decoding pipeline.
We adopt a two-dimensional tensor-network (TN) ansatz to simulate variational quantum algorithms on two-dimensional qubit architectures, demonstrating its capability to accurately simulate deep circuits through the Quantum Approximate Optimization Algorithm (QAOA) applied to Ising spin-glass problems on heavy-hexagonal and square lattices. For heavy-hexagonal problems with up to three-body interactions, parameters trained on small instances and transferred to systems an order of magnitude larger improve the sampled energy distribution only up to intermediate depths, indicating a fundamental limit of parameter concentration as a transfer strategy. By extending the training itself with TN simulations on larger system sizes, we avoid local minima and obtain lower-energy samples. Analyses of entanglement growth and importance sampling show that the simulation remains classically feasible with moderate bond dimension. We find that parameter concentration also persists on square lattices, albeit at substantially higher computational cost to perform reliable sampling. Overall, our TN framework not only provides an efficient and controlled framework for benchmarking variational quantum algorithms on two-dimensional lattices, but also serves as an effective surrogate model for training variational algorithms.
Efficient simulation of quantum circuits has become indispensable with the rapid development of quantum hardware. The primary simulation methods are based on state vectors and tensor networks. As the number of qubits and quantum gates grows larger in current quantum devices, traditional state-vector-based quantum circuit simulation methods prove inadequate due to the overwhelming size of the Hilbert space and extensive entanglement. Consequently, brutal force tensor network simulation algorithms become the only viable solution in such scenarios. The two main challenges faced in tensor network simulation algorithms are optimal contraction path finding and efficient execution on modern computing devices, with the latter determining the actual efficiency. In this study, we investigate the optimization of such tensor network simulations on modern GPUs and propose general optimization strategies from two aspects: computational efficiency and accuracy. First, we propose to transform critical Einstein summation operations into GEMM operations, leveraging the specific features of tensor network simulations to amplify the efficiency of GPUs. Second, by analyzing the data characteristics of quantum circuits, we employ extended precision to ensure the accuracy of simulation results and mixed precision to fully exploit the potential of GPUs, resulting in faster and more precise simulations. Our numerical experiments demonstrate that our approach can achieve a 3.96× reduction in verification time for random quantum circuit samples in the 18-cycle case of Sycamore, with sustained performance exceeding 21 TFLOPS on one A100. This method can be easily extended to the 20-cycle case, maintaining the same performance, accelerating by 12.5× compared to the state-of-the-art CPU-based results and 4.48–6.78× compared to the state-of-the-art GPU-based results reported in the literature. Furthermore, the strategies presented in this article hold general applicability for accelerating problems that can be tackled by contracting tensor networks such as graphical models and combinatorial optimizations.