A Project Submitted to the Graduate Faculty of the State University of New York Polytechnic Institute in Partial Fulfillment of the Requirements for the Degree of Master of Science Utica, New York
This project compares classical, quantum, and hybrid neural network architectures for multiclass classification on the MNIST and CIFAR10 datasets. The models include a classical convolutional network, two pure quantum neural networks constructed with TorchQuantum, a quanvolutional model that applied fixed random quantum filters to image patches, and hybrid quantum classical networks that combine classical convolution with trainable quantum layers. In addition, a quantum classical quantum convolutional neural network (QCQ-CNN) inspired by recent works is implemented, as well as a separate Qiskit based QCQ-CNN that uses the ZZFeatureMap for data encoding and the RealAmplitudes ansatz for variational processing. Most models were trained under shared hyperparameters and full multiclass supervision, while the Qiskit based QCQ-CNN was trained on a reduced MNIST and CIFAR10 subset in a One-vs-All setting due to the computational cost of simulating quantum circuits. Across both datasets, the classical convolutional neural network achieved the highest accuracy. The pure quantum models performed well and showed small gaps between training and testing accuracy, which suggests that the models did not overfit and were stable. The quanvolutional model produced expressive features but heavily overfit. The TorchQuantum QCQ-CNN achieved lower accuracy than the classical baseline but showed a consistent generalization and avoided the overfitting seen in the quanvolutional mode. The One-vs-All v QCQ-CNN had a strong accuracy for each class, showing that hybrid neural networks can perform well when the task is reduced into smaller subproblems. These results highlight the strengths and weaknesses of current quantum and hybrid neural networks in practical scenarios. The experiments illustrate that shallow quantum circuits can be integrated into classical neural networks to assist in their generalization ability even while the quantum aspects are simulated.
This project compares classical, quantum, and hybrid neural network architectures for multiclass classification on the MNIST and CIFAR10 datasets. The models include a classical convolutional network, two pure quantum neural networks constructed with TorchQuantum, a quanvolutional model that applied fixed random quantum filters to image patches, and hybrid quantum classical networks that combine classical convolution with trainable quantum layers. In addition, a quantum classical quantum convolutional neural network (QCQ-CNN) inspired by recent works is implemented, as well as a separate Qiskit based QCQ-CNN that uses the ZZFeatureMap for data encoding and the RealAmplitudes ansatz for variational processing. Most models were trained under shared hyperparameters and full multiclass supervision, while the Qiskit based QCQ-CNN was trained on a reduced MNIST and CIFAR10 subset in a One-vs-All setting due to the computational cost of simulating quantum circuits. Across both datasets, the classical convolutional neural network achieved the highest accuracy. The pure quantum models performed well and showed small gaps between training and testing accuracy, which suggests that the models did not overfit and were stable. The quanvolutional model produced expressive features but heavily overfit. The TorchQuantum QCQ-CNN achieved lower accuracy than the classical baseline but showed a consistent generalization and avoided the overfitting seen in the quanvolutional mode. The One-vs-All v QCQ-CNN had a strong accuracy for each class, showing that hybrid neural networks can perform well when the task is reduced into smaller subproblems. These results highlight the strengths and weaknesses of current quantum and hybrid neural networks in practical scenarios. The experiments illustrate that shallow quantum circuits can be integrated into classical neural networks to assist in their generalization ability even while the quantum aspects are simulated.
Deep learning has achieved significant success in pattern recognition, with convolutional neural networks (CNNs) serving as a foundational architecture for extracting spatial features from images. Quantum computing provides an alternative computational framework, a hybrid quantum-classical convolutional neural networks (QCCNNs) leverage high-dimensional Hilbert spaces and entanglement to surpass classical CNNs in image classification accuracy under comparable architectures. Despite performance improvements, QCCNNs typically use fixed quantum layers without incorporating trainable quantum parameters. This limits their ability to capture non-linear quantum representations and separates the model from the potential advantages of expressive quantum learning. In this work, we present a hybrid quantum-classical-quantum convolutional neural network (QCQ-CNN) that incorporates a quantum convolutional filter, a shallow classical CNN, and a trainable variational quantum classifier. This architecture aims to enhance the expressivity of decision boundaries in image classification tasks by introducing tunable quantum parameters into the end-to-end learning process. Through a series of small-sample experiments on MNIST, F-MNIST, and MRI tumor datasets, QCQ-CNN demonstrates competitive accuracy and convergence behavior compared to classical and hybrid baselines. We further analyze the effect of ansatz depth and find that moderate-depth quantum circuits can improve learning stability without introducing excessive complexity. Additionally, simulations incorporating depolarizing noise and finite sampling shots suggest that QCQ-CNN maintains a certain degree of robustness under realistic quantum noise conditions. While our results are currently limited to simulations with small-scale quantum circuits, the proposed approach offers a potentially promising direction for hybrid quantum learning in near-term applications.
In the last decade, quantum computing (QC) has shown its great potential in advancing research in many fields. Here we introduce a new quantum-classical neural network, by combining quantum computing and classical computing in a hybrid neural network (HNN) that can be trained to perform image classification. The HNN on the basis of the classic convolutional neural network (CNN) with quantum circuit is considered for image classification problem. The various configurations of HNN were investigated where QC with different number of qubits were used and compared. The HNN configurations were trained, validated, and tested on the more complex CIFAR10 and CIFAR100 datasets in addition to our previous attempts on the simpler MNIST, notMNIST, MNIST Fashion datasets. Performance of HNN was compared for multiclass classification on these datasets for different number of classes (from 2 to 10) using QCs with correspondent number of qubits (from 2 to 4). The metrics measured (accuracy and loss) during these experiments support our assumption about feasibility of HNN application for multiclass classification problems.
This study presents a systematic comparison between hybrid quantum-classical neural networks and purely classical models across three benchmark datasets (MNIST, CIFAR100, and STL10) to evaluate their performance, efficiency, and robustness. The hybrid models integrate parameterized quantum circuits with classical deep learning architectures, while the classical counterparts use conventional convolutional neural networks (CNNs). Experiments were conducted over 50 training epochs for each dataset, with evaluations on validation accuracy, test accuracy, training time, computational resource usage, and adversarial robustness (tested with ε=0.1 perturbations).Key findings demonstrate that hybrid models consistently outperform classical models in final accuracy, achieving {99.38\% (MNIST), 41.69\% (CIFAR100), and 74.05\% (STL10) validation accuracy, compared to classical benchmarks of 98.21\%, 32.25\%, and 63.76\%, respectively. Notably, the hybrid advantage scales with dataset complexity, showing the most significant gains on CIFAR100 (+9.44\%) and STL10 (+10.29\%). Hybrid models also train 5--12× faster (e.g., 21.23s vs. 108.44s per epoch on MNIST) and use 6--32\% fewer parameters} while maintaining superior generalization to unseen test data.Adversarial robustness tests reveal that hybrid models are significantly more resilient on simpler datasets (e.g., 45.27\% robust accuracy on MNIST vs. 10.80\% for classical) but show comparable fragility on complex datasets like CIFAR100 ($\sim$1\% robustness for both). Resource efficiency analyses indicate that hybrid models consume less memory (4--5GB vs. 5--6GB for classical) and lower CPU utilization (9.5\% vs. 23.2\% on average).These results suggest that hybrid quantum-classical architectures offer compelling advantages in accuracy, training efficiency, and parameter scalability, particularly for complex vision tasks.
Abstract In this study, we compare the performance of randomized classical and quantum neural networks (NNs) as well as classical and quantum-classical hybrid convolutional neural networks (CNNs) for the task of supervised binary image classification. We keep the employed quantum circuits compatible with near-term quantum devices and use two distinct methodologies: applying randomized NNs on dimensionality-reduced data, and applying CNNs to full image data. We evaluate these approaches on three fully-classical data sets of increasing complexity: an artificial hypercube data set, MNIST handwritten digits and industrial images of practical relevance. Our study’s central goal is to shed more light on how quantum and classical models perform for various binary classification tasks and on what defines a good quantum model. To this end, our study involves a correlation analysis between classification accuracy and quantum model hyperparameters, and an analysis on the role of entanglement in quantum models, as well as on the impact of initial training parameters. We find classical and quantum-classical hybrid models achieve statistically-equivalent classification accuracies across most data sets with no approach consistently outperforming the other. Interestingly, we observe that quantum NNs show lower variance with respect to initial training parameters and that the role of entanglement is nuanced. While incorporating entangling gates seems to be generally advantageous, we also observe that their (optimizable) entangling power is not correlated with model performance. We also observe an inverse proportionality between the number of entangling gates and the average gate entangling power. Our study provides an industry perspective on quantum machine learning for practical binary image classification tasks, highlighting both current limitations and potential avenues for further research in quantum circuit design, entanglement utilization, and model transferability across varied applications.
Convolutional Neural Networks (CNNs) are extensively utilized for image classification due to their ability to exploit data correlations effectively. However, traditional CNNs encounter several challenges, including limitations in accuracy improvement, computational efficiency, explain ability, and generalization. As data dimensionality or model complexity increases, CNNs often become more difficult to train efficiently, leading to slower processing times. To overcome these challenges, Quantum Convolutional Neural Networks (QCNNs) offer a promising quantum-based approach to enhance existing models, particularly by improving training efficiency and reducing processing times. This paper provides a comparative analysis of classical CNNs and a novel quantum circuit architecture tailored for image-based tasks. The analysis emphasizes the adaptability and potential of quantum circuits to enhance feature extraction, recognition accuracy, and processing efficiency.We applied various QCNN architectures to the Chinese-MNIST dataset and compared their performance with traditional CNN models. Although QCNNs achieved slightly lower accuracy, they demonstrated significantly greater stability. These findings high- light QCNNs’ potential to overcome the computational limitations of conventional deep learning, offering a promising pathway for quantum-enhanced solutions in computer vision. Future work will focus on refining quantum circuit designs, developing hybrid quantum-classical models, and extending QCNN applications to larger and more diverse datasets.
The objective of this research is to develop a Hybrid Quantum-Classical Convolutional Neural Network (QC-CNN) framework that integrates parameterized quantum circuits (PQCs) within classical CNN architectures to enhance image classification performance.The proposed model leverages quantum principles such as superposition and entanglement for high-dimensional feature representation, achieving superior accuracy and reduced computational complexity.Experiments on MNIST and CIFAR-10 datasets demonstrated improved classification accuracy of 98.7% and 82.5%, respectively, surpassing traditional CNNs while reducing training time by 33% and parameters by 45%.Statistical analysis confirmed the significance of these improvements.Visualizations using t-SNE revealed enhanced class separability, and noise perturbation tests validated the model's robustness.The results highlight the hybrid QC-CNN's potential for efficient and scalable quantum-enhanced deep learning applications.Extended evaluations, including quantum-layer-depth analysis, robustness testing, and t-SNE visualization, empirically support the hybrid QC-CNN's effectiveness across varied scenarios.