Performance benchmarking ofhybrid quantum-classical neuralnetworks on IBM hardware : Ansatz Architecture, Optimisation Behaviour, and Hard-ware Evaluation for Near-Term Quantum Classification
Variational quantum classifiers combine parameterised quantum circuits with classical optimisa-tion to perform supervised classification, but their practical performance depends strongly on circuit architecture, trainability, and hardware-resource requirements. Current noisy intermediate-scalequantum devices further constrain these models through limited usable circuit depth, hardwarenoise, and expensive quantum-circuit evaluations, motivating systematic benchmarking undercontrolled conditions.This thesis investigates how the structure of an interaction-based ansatz influences the behaviourof a hybrid quantum–classical classifier. A ZZ-only IsingZZ ansatz is compared with the proposed Arivoli ansatz, which combines XX-, Y Y -, and ZZ-type interaction channels between neigh-bouring qubits. The primary architecture comparison uses IsingZZ with three repeated nearest-neighbour ladders and Arivoli with one ladder, with nine nominal optimisation parameters in each implementation.Classical data are reduced to four features and encoded on four qubits using arcsin-RY rotations.The resulting quantum states are processed by the parameterised ansatz and measured usingAll-Qubit Multi-Observable Measurement (AMM), producing X-, Y -, and Z-expectation valuesthat are classified by a linear softmax readout. The architectures are evaluated on benchmark classification datasets using predictive performance, optimisation behaviour, expressibility, en-tangling capability, and circuit-resource metrics. Classical baselines and IBM Quantum execution are additionally used to assess the practical relevance and hardware behaviour of the proposedframework.The study does not seek to demonstrate quantum advantage. Instead, it examines the trade-offs between representational capability, optimisation behaviour, classification performance, and hard-ware cost in variational quantum classifier design. The results are used to assess whether richer multi-axis interactions provide practically useful behaviour under near-term quantum-computingconstraints.
Variational quantum classifiers combine parameterised quantum circuits with classical optimisa-tion to perform supervised classification, but their practical performance depends strongly on circuit architecture, trainability, and hardware-resource requirements. Current noisy intermediate-scalequantum devices further constrain these models through limited usable circuit depth, hardwarenoise, and expensive quantum-circuit evaluations, motivating systematic benchmarking undercontrolled conditions.This thesis investigates how the structure of an interaction-based ansatz influences the behaviourof a hybrid quantum–classical classifier. A ZZ-only IsingZZ ansatz is compared with the proposed Arivoli ansatz, which combines XX-, Y Y -, and ZZ-type interaction channels between neigh-bouring qubits. The primary architecture comparison uses IsingZZ with three repeated nearest-neighbour ladders and Arivoli with one ladder, with nine nominal optimisation parameters in each implementation.Classical data are reduced to four features and encoded on four qubits using arcsin-RY rotations.The resulting quantum states are processed by the parameterised ansatz and measured usingAll-Qubit Multi-Observable Measurement (AMM), producing X-, Y -, and Z-expectation valuesthat are classified by a linear softmax readout. The architectures are evaluated on benchmark classification datasets using predictive performance, optimisation behaviour, expressibility, en-tangling capability, and circuit-resource metrics. Classical baselines and IBM Quantum execution are additionally used to assess the practical relevance and hardware behaviour of the proposedframework.The study does not seek to demonstrate quantum advantage. Instead, it examines the trade-offs between representational capability, optimisation behaviour, classification performance, and hard-ware cost in variational quantum classifier design. The results are used to assess whether richer multi-axis interactions provide practically useful behaviour under near-term quantum-computingconstraints.
This study investigates classification performance and trainability of variational quantum classifiers on near-term noisy quantum hardware across circuit depth and qubit count regimes within the context of quantum computing and machine learning, an area of growing scientific importance given its implications for hybrid quantum-classical algorithm design and practical NISQ hardware benchmarking for near-term quantum advantage assessment. Using gradient-based parameter optimization of parameterized quantum circuits with hardware noise modeling via density matrix simulation and physical execution on IBM Eagle processors, we examine expressibility of parameterized circuits enabling nonlinear decision boundaries in feature Hilbert space, with trainability constrained by barren plateau gradients at large qubit counts in 96 circuit architectures (4 qubit counts x 4 depths x 6 ansatz families) evaluated across 5 classification datasets with 20 random initialization seeds each drawn from IBM Quantum Eagle (127-qubit) and Falcon (27-qubit) processors with noise mitigation via zero-noise extrapolation. Results indicate that 10-qubit, 4-layer VQC achieves 84.2% classification accuracy on the best benchmark, within 3.8 percentage points of classical SVM, but gradient variance decays exponentially with qubit count (variance halving per 2 added qubits), confirming barren plateau scaling (p < 0.001), with 84.2% accuracy, 3.8 pp gap to classical SVM baseline as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to quantum computing and machine learning and carry actionable implications for the design of programs and policies targeting hybrid quantum-classical algorithm design and practical NISQ hardware benchmarking for near-term quantum advantage assessment.
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.
uantum computing offers the potential to solve selected computational problems more efficiently than classical systems by leveraging quantum mechanical principles such as superposition and entanglement. However, current quantum devices operate in the noisy intermediatescale quantum (NISQ) regime, where limited coherence times, gate errors, measurement instability, and hardware variability significantly constrain practical performance. As a result, many quantum machine learning architectures that appear promising under idealized simulation conditions experience substantial degradation when executed under realistic noisy environments. This thesis investigates the problem of robustness in hybrid quantum neural network (HQNN) systems operating under realistic NISQ conditions. Rather than focusing primarily on quantum expressivity or claims of quantum advantage, the work examines whether practical hybrid quantum learning can maintain stable optimization and reliable inference in the presence of stochastic perturbations. A robustness-engineering framework is developed that integrates learned-readout decoding, multi-observable quantum feature extraction, architecture-level optimization, stability-aware training, and noise-augmented learning within a unified hybrid quantum-classical pipeline. To evaluate the framework, a comprehensive benchmarking methodology is developed across multiple datasets, architectures, perturbation environments, software frameworks, and hardware platforms. The experiments compare classical machine learning models, quantum kernel methods, and several HQNN architectures, including quanvolutional neural networks, quantum convolutional neural networks, and quantum transfer learning systems. Performance is evaluated under both noiseless and noisy conditions using standardized metrics, cross-noise validation, architecture-level comparisons, and real IBM Quantum hardware execution. The results demonstrate that baseline HQNN systems are highly sensitive to optimization instability, parity-based information loss, measurement variability, and perturbation accumulation. The experiments further reveal that robustness, rather than representational expressivity alone, is the dominant factor limiting practical hybrid quantum learning under realistic NISQ conditions. Through successive robustness-engineering stages, noisy performance improves from approximately 0.37 to 0.50 in baseline architectures to approximately 0.97 under several benchmark environments while maintaining substantially improved stability and reduced degradation under perturbation. These findings establish robustness as a central design objective for practical quantum machine learning. The thesis demonstrates that reliable hybrid quantum learning emerges through coordinated interaction between quantum feature generation, adaptive classical inference, stability-aware optimization, and hardware-aware architecture design. More broadly, the results suggest that hybrid quantum-AI systems are most effective when viewed as integrated quantum-classical computational frameworks in which quantum circuits function as probabilistic feature generators operating within larger robustness-oriented learning pipelines.
This thesis presents a comparative study of quantum machine learning and classical machine learning developed from first principles. It begins by establishing the foundations of the quantum circuit model, observables, and projective measurement, and then revisits the classical learning framework with an emphasis on risk minimisation, task-aligned metrics, and rigorous evaluation protocols. Building on this shared perspective, the thesis examines two central paradigms in quantum machine learning: quantum kernels and variational quantum circuits. A unifying theme is that quantum models must be assessed fairly against strong classical baselines and under realistic hardware constraints. The work therefore explicitly accounts for the limitations of noisy intermediate-scale quantum (NISQ) devices, including noise, restricted connectivity, compilation overhead, and finite-shot estimation. The empirical study instantiates both quantum pipelines on a controlled supervised benchmark adapted from IBM Quantum: classifying small synthetic images containing either a vertical or a horizontal line under bounded noise. Using the same dataset, identical splitting strategies, and matched budgets for shots and optimisation, the thesis compares quantum and classical models under a consistent evaluation protocol. The results show that both quantum kernel and variational approaches achieve meaningful accuracy in this regime, but do not demonstrate a decisive quantum advantage on the selected task. The thesis concludes by outlining limitations and directions for future work, including scaling to richer benchmarks, incorporating error mitigation, and strengthening assumption-explicit methodology for credible claims of quantum-enhanced learning.
Noisy Intermediate-Scale Quantum (NISQ) devices have led to the creation of hybrid quantum-classical algorithms, which have provided an innovative design to improve classical machine learning. This article focuses on a review of Hybrid Quantum-Classical Neural Networks (H-QCNNs) in image classification, which is an emerging area that attempts to exploit the representational ability of parameterized quantum circuits (PQCs) and the proven feature extraction capabilities of classical deep learning. We empirically examine the current tendencies in architecture, such as quanvolutional networks, sequential models, and variational quantum classifiers, and critically assess their performance on standard and custom data. The literature analysis shows a large performance difference between noiseless simulation and execution on real hardware, indicating the highly problematic aspects of scaling beyond binary tasks, quantum noise, and a lack of standardized benchmarking. Moreover, we mention such innovative contributions as the small H-QNN model that offers comparable accuracy, yet meets strict NISQ requirements. Last but not least, the current review offers a systematic taxonomy of H-QCNN architectures, a comparative study of methodology options, and a clear roadmap to future research to reach a workable quantum advantage in computer vision. It is a resource to the beginners as well as a reference to the professionals working in the field, so the field would be moved to more powerful, scalable, and hardware-conscious ways of implementation.
Quantum transfer learning combines pretrained classical deep learning models with quantum circuits to reuse expressive feature representations while limiting the number of trainable parameters. In this work, we introduce a family of compact quantum transfer learning architectures that attach variational quantum classifiers to frozen convolutional backbones for image classification. We instantiate and evaluate several classical-quantum hybrid models implemented in PennyLane and Qiskit, and systematically compare them with a classical transfer-learning baseline across heterogeneous image datasets. To ensure a realistic assessment, we evaluate all approaches under both ideal simulation and noisy emulation using noise models calibrated from IBM quantum hardware specifications, as well as on real IBM quantum hardware. Experimental results show that the proposed quantum transfer learning architectures achieve competitive and, in several cases, superior accuracy while consistently reducing training time and energy consumption relative to the classical baseline. Among the evaluated approaches, PennyLane-based implementations provide the most favorable trade-off between accuracy and computational efficiency, suggesting that hybrid quantum transfer learning can offer practical benefits in realistic NISQ era settings when feature extraction remains classical.