Fine-grained image classification under limited labeled data presents a representation challenge: pretrained deep networks provide rich but high-dimensional features, whereas small-data learning and practical quantum encoding require substantially more compact representations. Direct dimensionality reduction can make such features computationally tractable but may not preserve the class-discriminative geometry needed for effective kernel learning. This study, therefore, examines whether supervised latent restructuring can produce a compact, class-aware, qubit-compatible representation that supports classical and quantum-kernel learning. EfficientNet-B0 embeddings were reduced from 1280 to 64 dimensions using PCA and subsequently restructured by LDA into an 11-dimensional Supervised Latent Restructuring (SLR) representation. SLR increased the held-out Silhouette coefficient from −0.0054 in PCA64 to 0.0490; however, predictive effects were model dependent: Random Forest and XGBoost improved, Linear SVM changed little, and RBF-SVM favored PCA64. A dimension-matched PCA11 control showed no consistent predictive advantage of SLR over PCA11 across the evaluated classifiers. Under the original 12-class taxonomy, the fixed quantum kernel remained below matched classical baselines. Under a broader three-group taxonomy, quantum-kernel performance improved with increasing training support, but RBF-SVM remained the strongest. $\gamma $ -only Quantum Kernel Alignment did not consistently outperform the fixed kernel and incurred up to $420\times $ greater runtime than the fixed kernel. Target-control and kernel-structure analyses showed that increased alignment produced little additional class-discriminative structure. Overall, the study shows that representation geometry, label granularity, numerical scaling, and adaptive feature scaling jointly influence small-data quantum-kernel learning; no quantum advantage was demonstrated.
Fine-grained image classification under limited labeled data presents a representation challenge: pretrained deep networks provide rich but high-dimensional features, whereas small-data learning and practical quantum encoding require substantially more compact representations. Direct dimensionality reduction can make such features computationally tractable but may not preserve the class-discriminative geometry needed for effective kernel learning. This study, therefore, examines whether supervised latent restructuring can produce a compact, class-aware, qubit-compatible representation that supports classical and quantum-kernel learning. EfficientNet-B0 embeddings were reduced from 1280 to 64 dimensions using PCA and subsequently restructured by LDA into an 11-dimensional Supervised Latent Restructuring (SLR) representation. SLR increased the held-out Silhouette coefficient from −0.0054 in PCA64 to 0.0490; however, predictive effects were model dependent: Random Forest and XGBoost improved, Linear SVM changed little, and RBF-SVM favored PCA64. A dimension-matched PCA11 control showed no consistent predictive advantage of SLR over PCA11 across the evaluated classifiers. Under the original 12-class taxonomy, the fixed quantum kernel remained below matched classical baselines. Under a broader three-group taxonomy, quantum-kernel performance improved with increasing training support, but RBF-SVM remained the strongest. γ -only Quantum Kernel Alignment did not consistently outperform the fixed kernel and incurred up to 420× greater runtime than the fixed kernel. Target-control and kernel-structure analyses showed that increased alignment produced little additional class-discriminative structure. Overall, the study shows that representation geometry, label granularity, numerical scaling, and adaptive feature scaling jointly influence small-data quantum-kernel learning; no quantum advantage was demonstrated.
High-dimensional biological data often exhibit a severe mismatch between feature dimensionality and sample size, making reliable classification difficult in extremely small-data regimes. In these settings, kernel methods can lose discriminative power when latent compression fails to preserve class-separating structure. We study this problem in fine-grained plant phenomics and propose a hybrid workflow that compresses 1280-dimensional deep image embeddings into a 64-dimensional PCA space and then restructures them into an 11-dimensional supervised latent space using Linear Discriminant Analysis (LDA), followed by GPU-accelerated Quantum Kernel Alignment (QKA) on NVIDIA L40S hardware. Empirically, supervised latent restructuring substantially improves the geometric separability of the compressed representation, increasing the Silhouette coefficient from 0.003 in the raw embedding space and -0.006 in PCA-64 to 0.197 in the supervised LDA-11 space. However, downstream classical evaluation reveals a clear compression trade-off: Linear SVM and XGBoost improve in the restructured latent space, whereas RBF-SVM and Random Forest degrade under the same 11-dimensional bottleneck. Under a constrained optimization budget, QKA in this regime remains challenging, indicating that latent geometry alone is not sufficient for strong trainable quantum performance. These findings position representation geometry as a central design variable in small-data quantum learning and expose the practical difficulty of recovering nonlinear discriminative structure from aggressively compressed biological representations.
High-dimensional biomedical learning frequently suffers from fragile generalization and weak explanatory value when latent representations are optimized for variance alone. This study proposes an interpretable pipeline that couples Quantum Principal Component Analysis (Quantum PCA) with quantum kernel similarity learning to produce low-dimensional structure that remains predictive, geometrically faithful, and biologically auditable. Experiments were conducted on a harmonized cohort of n = 1038 samples with d = 2048 features across C = 3 diagnostic classes and 7 batch partitions. Compared with Classical PCA + SVM, Quantum PCA + SVM improved Macro-F1 from 0.821 to 0.848 and AUROC from 0.889 to 0.910, while reducing ECE from 0.044 to 0.038. Quantum Kernel SVM achieved the strongest discrimination with Macro-F1 0.868 and AUROC 0.923, while maintaining controlled calibration (ECE 0.039). Embedding diagnostics showed improved neighborhood preservation, with trustworthiness T(15) increasing from 0.916 under classical embeddings to 0.927 under Quantum PCA and 0.934 under quantum kernel geometry. Interpretability audits demonstrated stable component semantics, with loading stability ICC reaching 0.86 for PC1 and pathway enrichment attaining median FDR-adjusted significance on the order of 10-4. Robustness analysis under depolarizing perturbations exhibited gradual degradation, with Macro-F1 decreasing from 0.874 to 0.832 as noise increased to (p=0.010), and finite-shot sensitivity showing diminishing returns beyond 4096 shots as kernel variance stabilized. These results indicate that combining Quantum PCA and quantum kernels yields interpretable dimensionality reduction that improves predictive validity while preserving biomedical structure under practical constraints.
In supervised quantum machine learning with classical data, classification performance critically depends on the geometry induced by quantum data embedding. Existing data-adaptive approaches rely on pairwise fidelity evaluations between samples, which scale as O(B2) for B samples and create a computational bottleneck in iterative embedding optimization. To address this issue, we propose a prototype-guided framework for efficient data-adaptive quantum embedding. Using class-prototype separation and sample-to-prototype compactness, we derive a measurable lower certificate for class-ensemble distinguishability and construct a prototype-based objective requiring only O(B) fidelity evaluations. We integrate this objective into an end-to-end pipeline that combines task-specific quantum circuit architecture optimization with continuous refinement of the data-embedding parameters. Across five benchmark datasets with up to 20 qubits, our method retains competitive quantum kernel support vector machine (SVM) performance while reducing fidelity-evaluation cost, with particularly strong gains over budget-matched pairwise optimization on T-MNIST. Geometric analyses examine how the learned embeddings relate to kernel alignment and classification margins, revealing dataset-dependent benefits and limitations. We further derive a prototype-based classifier in the reproducing kernel Hilbert space that provides an explicit witness for the SVM objective and a low-cost proxy for model selection, and extend the framework to multiple prototypes for multimodal class structures.
Quantum kernel methods are often proposed as a route to improved medical image classification because quantum feature maps can embed data into high-dimensional Hilbert spaces. However, their practical value remains unclear when compared with strong classical kernels under leakage-controlled conditions. This study benchmarks two simulated quantum kernel support vector machines, QSVM-ZZ and QSVM-Pauli, against logistic regression, linear SVM, fixed-default RBF-SVM, and a compact multi-layer perceptron on BrainTumorMRI and BreastMNIST. All models used identical frozen ResNet18 features, train-only standardisation, PCA compression to four and eight dimensions, three random seeds, and matched classifier-training budgets where applicable. Across 24 seed-level matched comparisons, the best quantum kernel model never outperformed the best classical model. The mean quantum-minus-classical macro-F1 gap was - 0.2230, with all 24 gaps negative. Fixed pairwise comparisons confirmed that RBF-SVM exceeded both quantum kernels individually. QSVMs also required approximately 50 × to more than 220 × higher runtime. These findings show that, for compressed ResNet18 medical image features under noiseless simulation, the tested QSVM-ZZ and QSVM-Pauli kernels did not provide a practical advantage over fixed-default RBF-SVM.
Large language and sentence-embedding models provide rich semantic representations, but their high dimensionality poses a challenge for near-term quantum machine learning (QML), where quantum circuits can process only a limited number of input features. We investigate a hybrid quantum-classical pipeline that transforms high-dimensional sentence embeddings into compact representations for variational quantum classification. The workflow combines a pretrained sentence-embedding model, dimensionality reduction, angle encoding, a variational quantum circuit (VQC), and a classical decision layer. We systematically compare principal component analysis (PCA), neighborhood components analysis (NCA), and linear discriminant analysis (LDA), covering both unsupervised and supervised dimensionality reduction. Using the TREC question-classification dataset, we study the relationship between representation dimensionality, information retention, qubit count, and classification performance. Preliminary PCA experiments reveal a strong information bottleneck: reducing 768-dimensional embeddings to 3, 4, 5, and 8 dimensions retains about 8.2%, 10.2%, 11.9%, and 16.4% of the variance, with corresponding classification accuracies of 50.3%, 51.2%, 57.9%, and 63.4%. In contrast, supervised reduction is substantially more efficient. LDA reaches 85.3% accuracy and NCA reaches 83.1% using only 5 dimensions, under a leakage-free cross-validation protocol, compared with 85.1% for a full 384-dimensional classical baseline. These results indicate that supervised dimensionality reduction can preserve task-relevant information far more effectively than variance-based compression, making compact representations a promising route toward practical hybrid quantum-classical NLP models.
This thesis presents a systematic empirical evaluation of quantum machine learning performance under noisy intermediate-scale quantum (NISQ) era constraints. Through 670 controlled experiments, it evaluated quantum kernel support vector machines and variational quantum classifiers against classical baselines on MNIST binary and multiclass classification tasks with systematic variation of problem difficulty, feature dimensionality (4, 8 qubits), and training set size (n ∈{100, 250, 400, 500, 2000, 4000}). Statistical rigor was ensured through five random seeds per condition and comprehensive significance testing. During the testing with binary datasets, classical methods (SVM, logistic regression, k-NN, neural networks) achieved 85.9% to 99.6% accuracy with training times under one second across all sample sizes. Quantum kernel methods utilizing ZZ feature maps yielded 70.5% to 96.0% accuracy, requiring 129 to 2,140 seconds training with quadratic sample size scaling. Larger scale quantum experiments (n = 4000 for both architectures, 8-qubit encoding with n = 2000 for quantum kernels) proved computationally impossible on available hardware leading to out of memory errors, highlighting practical NISQ-era scalability constraints. Variational quantum classifiers exhibited persistent performance stagnation, failing to exceed 66.5% accuracy despite increases in training data and feature dimensionality. Performance gaps ranged from 3.3 to 35.2 percentage points. All quantum classical differences achieved statistical significance (p< 0.05, with 13 of 14 comparisons meeting p< 0.01) and exhibited large to very large efect sizes (d> 1.0, with 13 of 14 comparisons exceeding d> 2.0). To further evaluate performance and scalability, a 4 class classification experiment was conducted following an identical experimental protocol, with training set sizes of n ∈{100, 250, 400}. Across 180 runs, classical baselines achieved mean accuracies between 87.2% and 95% across both 4 and 8 dimensional data. Quantum Kernel SVM struggled during runs with 4 dimensions, accuracy varying between 64.6% and 74%. Quantum Kernel SVM accuracy declined 16%-25% when the dimension and qubit count were increased to 8, showing findings of exponential kernel concentration. The Variational Quantum Classifier also performed poorly, with accuracy ranging from 43.4% to 44.6% across all runs, providing further evidence that it failed to leverage additional features or training data. Experiments employed noiseless statevector simulation, establishing that quantum underperformance stems from fundamental algorithmic limitations independent of hardware noise. A further extension to 8 class classification (digits 0–7) confirmed and amplified these findings: classical methods maintained 65–83% accuracy while Quantum Kernel SVM degraded to 34–47% and VQC collapsed to 21–23%, approaching the 12.5% random baseline. Supplementary appendix experiments further examined amplitude encoding on raw 784 dimensional pixel features and an alternative angle encoding scheme on the 8-class task; the latter demonstrated that replacing the ZZ Feature Map with simple per qubit rotations substantially recovered quantum kernel performance, suggesting feature map design as a more consequential variable than circuit depth or qubit count in near-term quantum classification.