arXiv:2605.19417

Towards Fair Benchmarking of Quantum Transfer Learning for Visual Classification

Nouhaila Innan, Saim Rehman, Muhammad Shafique

On Aqora:

quant-phSubmitted May 19, 2026• Announced May 20, 2026• 0 citations

Abstract

Quantum Transfer Learning (QTL) offers a promising approach for visual quantum machine learning under near-term constraints, where limited qubit counts, shallow circuit depths, and costly hybrid optimization restrict end-to-end quantum training. In this setting, pretrained classical backbones can extract high-level visual features, while compact quantum modules operate as trainable classification heads. However, existing QTL results are difficult to compare because they often differ in datasets, preprocessing, backbone settings, qubit budgets, circuit designs, optimization choices, and reporting protocols. This work presents a controlled benchmarking methodology for evaluating representative QTL methods under a unified transfer-learning pipeline. The benchmark compares DQN-QTL, QPIE-QTL, AE-CQTL, PVCQTL, and ED-QTL under shared preprocessing rules, frozen-backbone settings, training conditions, and reporting metrics. The evaluation focuses on Fashion-MNIST and Hymenoptera Ants vs Bees as the two main datasets, while CIFAR-10 is used to provide additional configuration-level evidence on a harder natural-image task. Beyond predictive performance, the benchmark analyzes circuit size, trainable parameters, quantum parameters, training time, and architectural sensitivity to qubit count and circuit depth. The results show that no single QTL family dominates across all settings: performance depends on the dataset, encoding strategy, circuit design, and computational cost. These findings highlight the need for resource-aware QTL evaluation and provide guidance for selecting hybrid quantum-classical transfer models under near-term resource constraints.

Comments: 11 pages, 5 figures

Related papers

Quantum Transfer Learning with Adversarial Robustness for Classification of High‐Resolution Image Datasets

Amena Khatun, Muhammad Usman

Abstract The application of quantum machine learning to large‐scale high‐resolution image datasets is not yet possible due to the limited number of qubits and relatively high level of noise in the current generation of quantum devices. In this work, this challenge is addressed by proposing a quantum transfer learning (QTL) architecture that integrates quantum variational circuits with a classical machine learning network pre‐trained on ImageNet dataset. Through a systematic set of simulations over a variety of image datasets such as Ants & Bees, CIFAR‐10, and Road Sign Detection, the superior performance of the QTL approach over classical and quantum machine learning without involving transfer learning is demonstrated. Furthermore, the adversarial robustness of QTL architecture with and without adversarial training is evaluated, confirming that our QTL method is adversarially robust against data manipulation attacks and outperforms classical methods.

arXiv:2401.17009Advanced Quantum TechnologiesSep 22, 202423 citations

Post-variational classical quantum transfer learning for binary classification

Kavitha Yogaraj, Brian Quanz, Tarun Vikas +2 more

We address the limitations of variational quantum circuits (VQCs) in hybrid classical-quantum transfer learning by introducing post-variational strategies, which reduce training overhead and mitigate optimization issues. Our approach Post Variational Classical Quantum Transfer Learning (PVCQTL) includes three designs: (1) modified observable construction, (2) a hybrid approach, and (3) a variational-post-variational combination. We evaluate these on pre-trained models (VGG19, ResNet50, ResNet18, MobileNet) for 4 and 8 qubits, with ResNet50 performing best in deepfake detection. Compared to classical models (MLP, ResNet50) and quantum baselines hybrid quantum classical neural network (HQCNN), classical-quantum transfer learning (CQTL). PVCQTL consistently achieves better accuracy. The modified observable variant reaches 85% accuracy for Deepfake dataset with lower computational cost. To evaluate generalizability, we tested PVCQTL on three additional binary classification datasets, observing improved accuracy on each. We conducted ablation studies to assess the effects of architectural choices on quantum component variations, including the choice of quantum gates, use of fixed ansatz circuits, and observable measurements. Robustness to input noise and sensitivity of the PVCQTL models were examined through ablation studies on learning rate, batch size, and number of qubits. These results demonstrate that PVCQTL offers a measurable improvement over traditional hybrid classical-quantum approaches.

Scientific ReportsJul 2, 20257 citations

Hybrid quantum transfer learning for supervised UAV aerial image classification

S. Abdel‐Khalek, Mariam Algarni, Suliman A. Alsuhibany +1 more

Unmanned Aerial Vehicles (UAVs) enable high-resolution remote-sensing for land-cover and object mapping, yet conventional deep learning (DL) models often remain sensitive to scale, viewpoint, and illumination variability in UAV imagery. This paper presents QTL-SAIC, a quantum transfer learning framework for supervised aerial image classification that integrates classical feature learning with quantum-enhanced decision making. A pre-trained Capsule Network (CapsNet) extracts compact, discriminative features, which are classified using a variational quantum neural network (QNN) within a hybrid quantum–classical pipeline. The quantum module is fully specified (qubit number, encoding, circuit architecture, measurements, optimization, and execution backend with shot settings) to support reproducibility. Hyperparameters are optimized via the Gannet Optimization Algorithm (GOA), and evaluation follows stratified splits, repeated multi-seed trials and statistical reporting, with class imbalance addressed through cost-sensitive and/or balanced sampling strategies. Results indicate that QTL-SAIC yields competitive and stable performance relative to resource-matched classical baselines. Although no formal quantum advantage is claimed, ablations on circuit depth and noise sensitivity delineate the conditions under which the quantum component contributes to predictive performance.

Alexandria Engineering JournalMay 1, 20262 citations

Hybrid Classical-Quantum Transfer Learning with Noisy Quantum Circuits

D. Martín-Pérez, F. Rodríguez-Díaz, D. Gutiérrez-Avilés +2 more

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.

quant-pharXiv:2603.16973Computer Modeling in Engineering & SciencesMar 18, 20261 citation

Classification of Personal Protective Equipment with Quantum Transfer Learning in Industrial Training

Andrea Lucia Sulla Valdivia, Cesar Pio Castillo Cáceres, Jose Alfredo Sulla-Torres

Compliance with Personal Protective Equipment (PPE) in mechanical training workshops is essential but complex to monitor consistently. This work presents a Quantum Transfer Learning (QTL) pipeline that couples a pre-trained classical feature extractor with a variational quantum classifier to recognize helmet-use conditions in industrial training footage. We target a three-class setting—correct helmet use, incorrect helmet use, and no helmet—using transfer features from a lightweight convolutional backbone and a Basic-Entangler-type parameterized quantum circuit optimized with the Adam optimizer. The dataset comprises images collected in a controlled mechanical-training environment, with standard splits and class imbalance typical of educational settings. A proof-of-concept evaluation on the QTL-Helmet subset reports a test accuracy of 71.45% after 50 epochs. Furthermore, a significant reduction in training time was observed when using GPUs compared to CPUs, demonstrating the efficiency of accelerated computing. The model shows the viability of the quantum-classical approach for classification tasks with moderate computational resources. While preliminary and below operational thresholds, the study surfaces concrete levers for improvement within the QTL paradigm: stronger data curation and augmentation tailored to PPE attributes, better class rebalancing, deeper classical backbones for richer embeddings, circuit-depth/ansatz search with regularization, and hybrid ensembling of multiple shallow quantum heads. We also outline deployment considerations for near-term quantum hardware via simulator-to-hardware transfer and batching strategies. These results establish a reproducible baseline for quantum-enhanced PPE recognition in industrial and educational contexts and delineate a roadmap to close the performance gap through data, architecture, and training refinements, bringing QTL closer to practical safety monitoring in Industry 4.0.

International Conference on Computers in Management and Business (ICCMB)Feb 27, 20260 citations

An Amplitude-Encoding-Based Classical-Quantum Transfer Learning framework: Outperforming Classical Methods in Image Recognition

Shouwei Hu, Li, Xi, Banyao Ruan +1 more

The classical-quantum transfer learning (CQTL) method is introduced to address the challenge of training large-scale, high-resolution image data on a limited number of qubits (ranging from tens to hundreds) in the current Noisy Intermediate-Scale quantum (NISQ) era. existing CQTL frameworks have been demonstrate quantum advantages with a small number of parameters (around 50), but the performance of quantum neural networks is sensitive to the number of parameters. Currently, there is a lack of exploration into larger-scale quantum circuits with more parameters. This paper proposes an amplitude-encoding-based classical-quantum transfer learning (AE-CQTL) framework, accompanied by an effective learning algorithm. The AE-CQTL framework multiplies the parameters of quantum circuits by using multi-layer ansatz. Based on the AE-CQTL framework, we designed and implemented two CQTL neural network models: Transfer learning Quantum Neural Network (TLQNN) and Transfer Learning Quantum Convolutional Neural Network (TLQCNN). Both models significantly expand the parameter capacity of quantum circuits, elevating the parameter scale from a few dozen to over one hundred parameters. In cross-experiments with three benchmark datasets (MNIST, Fashion-MNIST and CIFAR10) and three source models (ResNet18, ResNet50 and DenseNet121), TLQNN and TLQCNN have exceeded the benchmark classical classifier in multiple performance metrics, including accuracy, convergence, stability, and generalization capability. Our work contributes to advancing the application of classical-quantum transfer learning on larger-scale quantum devices in future.

arXiv:2502.20184Feb 27, 20250 citations