Abstract When applying quantum computing to machine learning tasks, one of the first considerations is the design of the quantum machine learning model itself. Conventionally, the design of quantum machine learning algorithms relies on the “quantisation” of classical learning algorithms, such as using quantum linear algebra to implement important subroutines of classical algorithms, if not the entire algorithm, seeking to achieve a quantum advantage through possible run-time accelerations brought by quantum computing. However, recent research has started questioning whether quantum advantage via speedup is the right goal for quantum machine learning (Schuld and Killoran 2022 PRX Quantum 3(3):030101.). Research also has been undertaken to exploit properties that are unique to quantum systems, such as quantum contextuality, to better design quantum machine learning models (Bowles et al. 2023). In this paper, we take an alternative approach by incorporating the heuristics and empirical evidences from the design of classical deep learning algorithms to the design of quantum neural networks. We first construct a model based on the data reuploading circuit (Pérez-Salinas et al. 2020 Quantum 4(226):226) with the quantum Hamiltonian data embedding unitary (Schuld and Petruccione 2021). Through numerical experiments on image datasets, including the famous MNIST and FashionMNIST datasets, we demonstrate that our model outperforms the quantum convolutional neural network (QCNN) (Cong et al. 2019 Nat Phys 15(12):1273–1278) by a large margin (up to over 40% on MNIST test set). Based on the model design process and numerical results, we then laid out six principles for designing quantum machine learning models, especially quantum neural networks.
Abstract When applying quantum computing to machine learning tasks, one of the first considerations is the design of the quantum machine learning model itself. Conventionally, the design of quantum machine learning algorithms relies on the “quantisation” of classical learning algorithms, such as using quantum linear algebra to implement important subroutines of classical algorithms, if not the entire algorithm, seeking to achieve a quantum advantage through possible run-time accelerations brought by quantum computing. However, recent research has started questioning whether quantum advantage via speedup is the right goal for quantum machine learning (Schuld and Killoran 2022 PRX Quantum 3(3):030101.). Research also has been undertaken to exploit properties that are unique to quantum systems, such as quantum contextuality, to better design quantum machine learning models (Bowles et al. 2023). In this paper, we take an alternative approach by incorporating the heuristics and empirical evidences from the design of classical deep learning algorithms to the design of quantum neural networks. We first construct a model based on the data reuploading circuit (Pérez-Salinas et al. 2020 Quantum 4(226):226) with the quantum Hamiltonian data embedding unitary (Schuld and Petruccione 2021). Through numerical experiments on image datasets, including the famous MNIST and FashionMNIST datasets, we demonstrate that our model outperforms the quantum convolutional neural network (QCNN) (Cong et al. 2019 Nat Phys 15(12):1273–1278) by a large margin (up to over 40% on MNIST test set). Based on the model design process and numerical results, we then laid out six principles for designing quantum machine learning models, especially quantum neural networks.
Artificial neural networks have achieved great success in many fields ranging from image recognition to video understanding. However, its high requirements for computing and memory resources have limited further development on processing big data with high dimensions. In recent years, advances in quantum computing show that building neural networks on quantum processors is a potential solution to this problem. In this paper, we propose a novel neural network model named Quantum Convolutional Neural Network (QCNN), aiming at utilizing the computing power of quantum systems to accelerate classical machine learning tasks. The designed QCNN is based on implementable quantum circuits and has a similar structure as classical convolutional neural networks. Numerical simulation results on the MNIST dataset demonstrate the effectiveness of our model.
Abstract Image classification, a pivotal task in multiple industries, faces computational challenges due to the burgeoning volume of visual data. This research addresses these challenges by introducing two quantum machine learning models that leverage the principles of quantum mechanics for effective computations. Our first model, a hybrid quantum neural network with parallel quantum circuits, enables the execution of computations even in the noisy intermediate-scale quantum era, where circuits with a large number of qubits are currently infeasible. This model demonstrated a record-breaking classification accuracy of 99.21% on the full MNIST dataset, surpassing the performance of known quantum–classical models, while having eight times fewer parameters than its classical counterpart. Also, the results of testing this hybrid model on a Medical MNIST (classification accuracy over 99%), and on CIFAR-10 (classification accuracy over 82%), can serve as evidence of the generalizability of the model and highlights the efficiency of quantum layers in distinguishing common features of input data. Our second model introduces a hybrid quantum neural network with a Quanvolutional layer, reducing image resolution via a convolution process. The model matches the performance of its classical counterpart, having four times fewer trainable parameters, and outperforms a classical model with equal weight parameters. These models represent advancements in quantum machine learning research and illuminate the path towards more accurate image classification systems.
In recent years, there has been a growing interest in leveraging the unique properties of quantum computing to develop novel machine learning algorithms and architectures. This research paper presents an investigation of quantum convolutional neural networks (QCNNs), which leverage the unique properties of quantum computing to potentially improve the accuracy and efficiency of image classification tasks. Specifically, the paper explores three different QCNN architectures, including a pure quantum-based QCNN, a hybrid QCNN with a single quantum convolution layer, and a hybrid convolutional architecture with multiple quantum filters. We tested the models on MNIST dataset and the results of the study demonstrate that hybrid architectures that combine quantum and classical processing are more effective than pure quantum-based architectures in image classification tasks. In particular, the third model, the Hybrid Convolution with Multiple Quantum Filters, achieved the highest test set accuracy of 92.7%. The use of multiple quantum filters in conjunction with a classical neural network resulted in enhanced accuracy and efficiency in image classification tasks, highlighting the potential of hybrid architectures for future applications in machine learning tasks.
Quantum computing is expected to fundamentally change computer systems in the future. Recently, a new research topic of quantum computing is the hybrid quantum–classical approach for machine learning, in which a parameterized quantum circuit, also called quantum neural network (QNN), is optimized by a classical computer. This hybrid approach can have the benefits of both quantum computing and classical machine learning methods. In this early stage, it is of crucial importance to understand the new characteristics of quantum neural networks for different machine learning tasks. In this paper, we will study quantum neural networks for the task of classifying images, which are high-dimensional spatial data. In contrast to previous evaluations of low-dimensional or scalar data, we will investigate the impacts of practical encoding types, circuit depth, bias term, and readout on classification performance on the popular MNIST image dataset. Various interesting findings on learning behaviors of different QNNs are obtained through experimental results. To the best of our knowledge, this is the first work that considers various QNN aspects for image data.
Quantum Computing (QC) was born as a purely theoretical subject. Thus, in the beginning, the main category of researchers in the area was physicists and mathematicians. Since IBM introduced the first quantum computer, global interest in this emerging technology has been growing more and more, involving computer science and engineering researchers. Machine learning is another rapidly growing technology, used in a wide range of applications where traditional computing is not sufficient. The two technologies are merged in the thesis work. The different ways in which a combination of them can occur are discussed, as well as the explanation why there is not a single definition of Quantum Machine Learning (QML). Firstly, quantum computing is examined in depth. The basic elements of quantum computation theory, based on the fundamental principles of quantum mechanics, are introduced. The attention is focused on the qubit description on the Bloch sphere, the fundamental quantum gates, and examples of quantum algorithms. Therefore, the differences between quantum simulators and quantum processors are analysed, including the frameworks that provide tools to interact with them. Focusing on Quantum Machine Learning, the different definitions of the technology are described. There are four distinct approaches that are differentiated by the type of data used (quantum or classical) and by the device processing them (quantum or classical). In this thesis, QML is defined as the approach that uses classical data to feed a quantum device. The two possible methods for developing QML algorithms are: 1) to translate existing classical models into the terms of quantum computing; 2) to leave the existing models of classical machine learning completely, and explore new possible ways of doing QML to take full advantage of the physical characteristics of a quantum computer. The strategy adopted is the first one, thus it is used a state-of-the-art hybrid model of Quanvolutional Neural Network (QNN), composed of both quanvolutional and classical convolutional layers. The quanvolutional layers reproduce the same concept of the classical ones: they are made of quantum filters that generate feature maps through locally transforming input data. The transformation can be performed by variational quantum circuits or random quantum circuits. The specific field where Quantum Machine Learning is applied is image classification. The considered case study is a labeled dataset with X-ray images of lungs (labels: pneumonia, normal). The images are pre-processed and cropped for being used by quantum circuits, thus some information is lost. The main aim of the thesis is to understand whether and in which ways Quantum Machine Learning could bring advantages in the mentioned classification in terms of timing, accuracy, or dataset reduction, despite the large decrease in images' resolution. Different configurations of the hybrid model (with a different number of quantum filters and different settings of convolutional layers) are analysed and tested. The current quantum devices have many limitations, thus the time is still immature to observe a significant change in perspective. Nevertheless, the obtained results are promising for the future of QML in terms of dataset and hyperparameters reduction.
The rapid growth of computer vision and increasingly complex image recognition tasks has exposed fundamental computational limitations of classical machine learning models, motivating the exploration of quantum computing as an emerging new paradigm. This paper presents a comprehensive benchmarking study of classical and quantum machine learning models for image recognition on the MNIST handwritten digit dataset, evaluating both traditional models, a Classical Support Vector Machine (CSVM) and a Quantum Support Vector Machine (QSVM), and deep neural network models, a Classical Convolutional Neural Network (CCNN) and a Quantum Convolutional Neural Network (QCNN), across four performance dimensions: classification accuracy, computational runtime, parameter count, and memory requirements. Experiments are conducted as functions of both feature dimensionality and sample size, and across CPU and GPU execution environments, providing a controlled, multidimensional comparison to address gaps in prior work. For the SVM-based models, QSVM consistently outperforms CSVM in accuracy, reaching ∼ 0.90 versus ∼ 0.85 at 1,000 samples, with a higher computational cost. A feature count of 10 qubits and a sample size in the range of 200 -- 500 emerge as practical operating points that balance accuracy and runtime. For the neural network models, CCNN and QCNN achieve comparable classification accuracy, both exceeding 0.96 at 64 features and 60,000 samples, yet QCNN offers superior parameter and memory efficiency at higher feature counts, while incurring higher runtime. Across both model families, quantum models consistently outperform classical models by greater margins in accuracy as feature dimensionality or sample size increases.