Findings establish QCBM as a viable complementary tool for data augmentation, particularly for low-dimensional structured tabular data with class imbalance, particularly for low-dimensional structured tabular data with class imbalance.
Abstract
Data scarcity and class imbalance are persistent challenges in machine learning that degrade model generalization and introduce predictive bias. We present a hybrid quantum-classical framework for synthetic data generation using a Quantum Circuit Born Machine (QCBM) to address these limitations. The proposed approach exploits quantum mechanical properties -- superposition and entanglement -- within a parameterized variational quantum circuit to model complex probability distributions that are difficult for classical generative methods to capture. Experiments are conducted on two tabular benchmark datasets: the Iris dataset and the Telco Customer Churn dataset. Preprocessing includes normalization and PCA-based dimensionality reduction to enable efficient basis encoding for quantum circuits. The QCBM is trained by minimizing Kullback-Leibler (KL) divergence between real and generated data distributions using a gradient-based parameter-shift optimization rule. Augmenting training data with QCBM-generated synthetic samples at 40-50% of the minority class improves F1-score by approximately 5-15% and minority-class recall by 10-25%. Cross-domain evaluations (Train on Synthetic, Test on Real; and Train on Real, Test on Synthetic) reveal a performance gap of only 3-10%, indicating strong distributional fidelity. Comparative analysis against classical oversampling methods -- SMOTE, Borderline-SMOTE, KMeansSMOTE, and SVM-SMOTE -- shows that QCBM achieves competitive classification performance and produces lower Maximum Mean Discrepancy (MMD) on the Telco dataset, suggesting superior structural similarity in certain imbalanced settings. These findings establish QCBM as a viable complementary tool for data augmentation, particularly for low-dimensional structured tabular data with class imbalance.
We present \emph{spectral Born machines}, a class of quantum generative models that results from viewing and generalizing the class of IQP Born machines through the lens of group Fourier analysis. These quantum models exploit the quantum Fourier transform to create an inductive bias that make them naturally suited to learning integer-structured data, while remaining classically hard to sample from in general. Similar to IQP Born machines, spectral Born machines can be trained efficiently at scale on classical hardware via a maximum mean discrepancy loss based on graph spectral analysis, which we make available in a new \emph{tcdq} module of the PennyLane software platform. In numerical experiments, we show how the spectral bias of the model leads to significantly reduced parameter counts compared to unstructured approaches, and demonstrate the scalability of the software by training a 190-qubit model with over 1 million parameters to successfully learn a distribution of 93 nucleotide-long ribosomal RNA. Our results suggest that highly over-parameterized spectral Born machines may be immune to overfitting, even in strongly data-scarce regimes.
Austin L. Huang, William Maxwell, Vasilis Belis et al.· 3 citations
Quantum machine learning (QML) faces practical limitations due to noisy intermediate-scale quantum (NISQ) constraints, including noise, restricted qubit availability, and unstable optimization. This paper proposes HyQNet, a resource-aware hybrid quantum–classical framework designed to address these challenges through efficient circuit execution and adaptive optimization. The framework integrates optimized quantum circuits with classical learning strategies to improve scalability and stability under NISQ conditions. Experimental results on Iris, Wine, and Breast Cancer datasets show that HyQNet achieves an accuracy of 95.1% and F1-score of 94.8%, outperforming variational QNN (92.6%) and quantum SVM (91.2%). It also reduces runtime to 16.9 s compared to 20.5 s for VQNN, while maintaining efficient utilization of 8 qubits. Statistical analysis confirms significance (p < 0.05), and ablation studies validate the contribution of each component. The results demonstrate improved convergence stability and resource efficiency in hybrid quantum learning systems.
Sudheer Reddy K., Hastimal Jangid, Usha Desai· 2026 International Conferenc...· 0 citations
Overall, the results show that quantum and hybrid quantum-classical generative models can learn non-trivial discrete probability distributions but that their effectiveness depends strongly on the selected quantum model, ansatz, and training objective.
Introduction: Variational quantum algorithms can show unstable trainability and uneven generalisation that are not fully explained by circuit depth or raw parameter count alone. We study the data quantum Fisher information matrix (DQFIM) as a data-dependent effective-capacity diagnostic for controlled supervised unitary-learning tasks.
Materials and methods: We used exact-state, classically simulated, 4-qubit, supervised unitary-learning benchmarks with two ansatz families: a hardware-efficient ansatz (HEA) for the baseline trainability study and a symmetry-preserving ansatz for symmetry-controlled comparisons. The analysis covered four linked experiments: E0 trainability phase boundaries, E1 symmetry-controlled data-regime comparisons, E2 trainability versus generalisation, and E3 pre-training prediction of optimisation success. For the main analyses we used the support-basis DQFIM, 30 random seeds per main configuration, grouped cross-validation in the predictive benchmark, and additional diagnostics for threshold sensitivity, phase indeterminacy, Hamming-sector generalisation, and test set size sensitivity.
Results: In E0, the empirical trainability boundary increased from
M
c
=
32
at
L
=
1
to
M
c
=
192
at
L
=
8
, and the support-basis DQFIM-predicted boundary matched the empirical boundary on the resolved scanned grid under the main analysis setting. In E1, the sector-preserving and sector-broken conditions showed no resolved empirical boundary shift on the scanned grid, with only mild low-L asymmetry in the DQFIM-predicted boundary. In E2, trainability and generalisation separated clearly: some regimes generalised well, while others reached near-zero training loss but retained high test loss. The phase diagnostic supported relative phase indeterminacy as a mechanism for failure on in-sector superpositions after basis-state training. In E3, parameter count alone was a weak predictor of optimisation success, with receiver operating characteristic area under the curve (ROC AUC)
=
0.648
; the structural baseline was stronger, with ROC AUC
=
0.958
; and the DQFIM-enhanced model performed best, with ROC AUC
=
0.989
.
Conclusions: In these small, idealised, classically simulated matched-family tasks, the support-basis DQFIM provides a useful data-dependent pre-training diagnostic of effective capacity on the retained data support. It tracks resolved trainability boundaries and contributes predictive information beyond raw parameter count and structural metadata. The conclusions remain scoped to controlled small-system simulations; larger-qubit, noisy, hardware-executed, and less-matched settings require further validation before claiming practical scalability.
Shreyosha Ganguly, A. Masta, Shalini Devendrababu et al.· Academia Quantum· 0 citations
Noisy Intermediate-Scale Quantum (NISQ) devices can be used to solve classification problems using a promising quantum machine learning technique called Variational Quantum Classifiers (VQC). The performance of VQCs is highly dependent on classical optimizers, and the understanding of the optimizer behavior within a given quantum architecture is limited. The proposed work is to present an Adaptive Optimizer-Enhanced Variational Quantum Classifier, which is a combination of a ZZFeatureMap for quantum data encoding and TwoLocal ansatz for variational learning, for solving non-linear binary classification problems. The proposed framework is tested using the Two Moons benchmark dataset, which is a popular benchmark for evaluating the ability of machine learning models to learn complex non-linear decision boundaries. The implementation is based on the Qiskit 1.2 and was run on a noiseless state vector simulator to remove the effects of hardware noise and focus on the effects of optimization. The three popular optimization techniques, namely ADAM, COBYLA and SPSA were explored under the same experimental setups. Besides the classification accuracy, a callback-based loss tracking mechanism was added to examine the speed of convergence, stability of optimization and evolution of losses during training. The experimental results showed that the overall accuracy of ADAM is 84.2%, which is the best, followed by COBYLA (79.3%) and SPSA (73.8%). Along with this, quantitative convergence analysis results showed that ADAM converged faster than the other optimizers and had lesser loss variance. It is shown that, with a proper optimization routine, a shallow two-qubit VQC can be used to successfully learn complex non-linear decision boundaries. The principal contribution of this work is a controlled optimizer-aware evaluation framework that integrates a fixed ZZFeatureMap–TwoLocal VQC architecture with callback-based convergence diagnostics to systematically analyse optimizer behaviour under identical experimental conditions.
R. D, Ramya R.R, Sridevi S et al.· International Research Journ...· 0 citations