Variational Quantum Eigensolver (VQE) is a leading algorithm for quantum chemistry on near-term devices, but its performance is often limited by inefficient parameter initialization, which leads to slow convergence and high computational overhead. While meta-learning approaches such as LSTM-VQE have been proposed to predict optimization trajectories, their applicability remains restricted by rigid ansatz structures and molecule-specific parameter dimensions. To address these limitations, we propose LSTM-FC-VQE, an enhanced meta-learning framework that combines Long Short-Term Memory (LSTM) networks with a fully connected projection layer. This design enables the model to standardize variable-length parameter inputs, allowing it to handle molecular systems with heterogeneous qubit counts and ansatz architectures more effectively. We validate our approach on benchmark molecules, including H4and H2O, and demonstrate that LSTM-FC-VQE significantly reduces the number of VQE iterations required for convergence while achieving energy accuracies comparable to full configuration interaction (FCI) results. Beyond accelerating convergence, our method generalizes to unseen molecular configurations with minimal training data, offering a scalable and data-efficient strategy for quantum chemical simulations in the NISQ era.
Variational Quantum Eigensolver (VQE) is a leading algorithm for quantum chemistry on near-term devices, but its performance is often limited by inefficient parameter initialization, which leads to slow convergence and high computational overhead. While meta-learning approaches such as LSTM-VQE have been proposed to predict optimization trajectories, their applicability remains restricted by rigid ansatz structures and molecule-specific parameter dimensions. To address these limitations, we propose LSTM-FC-VQE, an enhanced meta-learning framework that combines Long Short-Term Memory (LSTM) networks with a fully connected projection layer. This design enables the model to standardize variable-length parameter inputs, allowing it to handle molecular systems with heterogeneous qubit counts and ansatz architectures more effectively. We validate our approach on benchmark molecules, including H4and H2O, and demonstrate that LSTM-FC-VQE significantly reduces the number of VQE iterations required for convergence while achieving energy accuracies comparable to full configuration interaction (FCI) results. Beyond accelerating convergence, our method generalizes to unseen molecular configurations with minimal training data, offering a scalable and data-efficient strategy for quantum chemical simulations in the NISQ era.
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.
Simulating strongly correlated molecular systems on near-term quantum hardware remains challenging because current hardware offers limited quantum volume and moderate-fidelity qubits. One way to address this challenge is through bootstrap embedding (BE). Bootstrap embedding partitions molecules into smaller fragments that are then embedded in the ``bath'' of other fragments iteratively. Bootstrap embedding is appealing for quantum simulation because fragmentation reduces the qubit requirements of each fragment problem. In this work, we develop a quantum bootstrap embedding (QBE) workflow that uses variational quantum eigensolver (VQE) fragment solvers and examines the algorithmic choices that determine the success of the overall VQE-QBE procedure. To improve efficiency, we introduce FastAdaptVQE, a sparse-matrix-accelerated form of the adaptive variational quantum eigensolver (ADAPT-VQE) that replaces symbolic commutator evaluation with direct statevector calculations, and MatrixFreeAdaptVQE, a matrix-free extension that removes the sparse-matrix memory bottleneck that arises for larger fragments. We also explore modifying the ADAPT-VQE operator-selection step by replacing the purely greedy choice with a lookahead strategy. Benchmarks on H4, F2, and selected geometries of H6 reach chemical accuracy, within 1 kcal/mol of BE results obtained with a full configuration interaction (FCI) solver. For n-butane, density-matching residuals converged, but energy errors remained outside chemical accuracy. These results show that combining QBE with VQE can accurately compute molecular energies, while density convergence alone does not ensure energy accuracy. This work lays the foundation for extending these energy calculations to larger molecular systems and quantum materials on near-term quantum hardware.
The Quantum Approximate Optimization Algorithm (QAOA) is a leading approach for solving combinatorial optimization problems on near-term quantum processors. However, finding good variational parameters remains a significant challenge due to the non-convex energy landscape, often resulting in slow convergence and poor solution quality. In this work, we propose a quantum meta-learning framework that trains advanced quantum sequence models to generate effective parameter initialization policies. We investigate four classical or quantum sequence models, including the Quantum Kernel-based Long Short-Term Memory (QK-LSTM), as learned optimizers in a "learning to learn" paradigm. Our numerical experiments on the Max-Cut problem demonstrate that the QK-LSTM optimizer achieves superior performance, obtaining the highest approximation ratios and exhibiting the fastest convergence rate across all tested problem sizes (n=10 to 13). Crucially, the QK-LSTM model achieves perfect parameter transferability by synthesizing a single, fixed set of near-optimal parameters, leading to a remarkable sustained acceleration of convergence even when generalizing to larger problems. This capability, enabled by the compact and expressive power of the quantum kernel architecture, underscores its effectiveness. The QK-LSTM, with only 43 trainable parameters, substantially outperforms the classical LSTM (56 parameters) and other quantum sequence models, establishing a robust pathway toward highly efficient parameter initialization for variational quantum algorithms in the NISQ era.
Quantum chemistry is one of the most promising applications of quantum computers in the near future. For noisy intermediate-scale quantum devices, the quantum-classical hybrid framework based on the variational quantum eigensolver (VQE) has become the method of choice. In the literature, there are many different variants of VQE, but it is not known which one is optimal for a given molecule. For this purpose, we perform a thorough benchmarking on more than ten different kinds of VQE ansatzes (in systems up to 30 qubits), based on their performance on the energy accuracy, runtime until convergence, and number of parameters. Our results show that the ADAPT ansatz can be used to obtain more accurate energy for small systems (below 14 qubits), but it costs much more computational resources. For larger molecules, UCCSD0 has better performance. However, all the tested ansatzes can hardly reach chemical accuracy at stretched bond lengths. Our results were obtained using MindSpore Quantum, where the codes and the benchmarking toolkit are publicly available at Gitee.
We present a novel quantum generative framework based on Quantum Long Short-Term Memory (QLSTM) networks to optimize the sampling of Slater determinants in Full Configuration Interaction (FCI), a key method in quantum chemistry for solving the Schrödinger equation. Our model employs variational quantum circuits to learn an approximate distribution over the configuration space, guiding the generation of new, physically meaningful determinants. This distribution reflects the correlations among significant electronic configurations, allowing for a guided exploration of the determinant space and accelerating convergence toward reference energies. By framing determinant selection as a generative modeling task, our approach contributes to the emerging field of quantum generative AI applied to many-body systems. Implemented using Pennylane and accelerated with JAX, the QLSTM model converges faster than both classical LSTM and random baselines. Notably, it exhibits stable and consistent convergence across different hyperparameter settings, suggesting that the model captures an underlying physical structure rather than overfitting to specific training conditions. Our results demonstrate the scalability and physical consistency of hybrid quantum-classical generative models for quantum chemistry tasks.
In Variational Quantum Algorithms (VQAs), circuit parameters are typically re-optimized from scratch for each new dataset, an approach that becomes inefficient in continuous computation settings where data arrive incrementally and statistical properties evolve over time. This limitation is exacerbated on noisy intermediate-scale quantum (NISQ) devices, where optimization is costly and prone to barren plateaus. To overcome these challenges, we introduce QURIOSO (QUantum gate parameteR predictIOn through quantum-enhanced long-Short term memOry), a machine learning framework that replaces repeated optimization with parameter prediction. QURIOSO learns trajectories of PQC parameters from past data and forecasts parameters for new data using either classical or quantum-enhanced LSTM architectures. We redesign quantum LSTM cells by employing Y-axis controlled rotations for amplitude modulation and full entanglement generation, enabling richer transformations than conventional phase-based designs. The framework operates in two modes: direct parameter transfer to unseen data and prediction-driven initialization to accelerate subsequent optimization. Validation on real-world binary classification tasks with hybrid classical–quantum models shows that quantum-enhanced LSTMs frequently outperform classical counterparts, highlighting their ability to preserve non-linearity through quantum encoding and measurement, and demonstrating the effectiveness of QURIOSO in generalizing across parameter trajectories within continuous quantum learning scenarios.