Quantum neural networks (QNNs), typically implemented via parameterized quantum circuits (PQCs), offer a potential route to quantum advantage in learning. We review recent progress through the Fourier structure of PQCs: the encoding Hamiltonian fixes the accessible frequency set, while training estimates the corresponding Fourier coefficients. This perspective explains how QNNs can mitigate the spectral bias that limits classical deep neural networks and motivates task-specialized designs. In particular, quantum ordinary differential equation (QODE) solver, quantum convolutional neural networks (QCNNs) and quantum reinforcement learning (QRL) exploit frequency pinching to better capture high-frequency image features and structured value landscapes. We further discuss trainability on noisy intermediate-scale quantum (NISQ) hardware, emphasizing Lie-algebraic and Fourier views of barren plateaus and practical mitigation strategies. Together, the Fourier and Lie perspectives yield concrete design principles and motivate quantum-native architectures built around spectral bias avoidance rather than classical imitation. Quantum neural networks (QNNs), implemented via parameterized quantum circuits (PQCs), promise a path to quantum advantage in learning by addressing the limitations of classical deep neural networks. Here, the authors explore the Fourier structure of PQCs to mitigate spectral bias, enhancing applications like quantum ODE solvers and quantum reinforcement learning, with implications for designing quantum-native architectures.
Quantum neural networks (QNNs), typically implemented via parameterized quantum circuits (PQCs), offer a potential route to quantum advantage in learning. We review recent progress through the Fourier structure of PQCs: the encoding Hamiltonian fixes the accessible frequency set, while training estimates the corresponding Fourier coefficients. This perspective explains how QNNs can mitigate the spectral bias that limits classical deep neural networks and motivates task-specialized designs. In particular, quantum ordinary differential equation (QODE) solver, quantum convolutional neural networks (QCNNs) and quantum reinforcement learning (QRL) exploit frequency pinching to better capture high-frequency image features and structured value landscapes. We further discuss trainability on noisy intermediate-scale quantum (NISQ) hardware, emphasizing Lie-algebraic and Fourier views of barren plateaus and practical mitigation strategies. Together, the Fourier and Lie perspectives yield concrete design principles and motivate quantum-native architectures built around spectral bias avoidance rather than classical imitation. Quantum neural networks (QNNs), implemented via parameterized quantum circuits (PQCs), promise a path to quantum advantage in learning by addressing the limitations of classical deep neural networks. Here, the authors explore the Fourier structure of PQCs to mitigate spectral bias, enhancing applications like quantum ODE solvers and quantum reinforcement learning, with implications for designing quantum-native architectures.
The mechanism governing the training dynamics of Quantum Neural Networks (QNNs) remains under-explored. In classical Deep Neural Networks (DNNs), training is dominated by "Spectral Bias," i.e. prioritizing learning low-frequency components and struggle for high-frequency details. In this work, we theoretically and empirically identify a distinct mechanism in QNNs, which we term Spectral Amplitude Priority. By analyzing the frequency-domain gradients and residual dynamics via the Quantum Neural Tangent Kernel (QNTK), we prove that QNN training is governed primarily by the magnitude of spectral components rather than their frequency indices. Consequently, QNNs can efficiently capture high-frequency functions-provided they have significant amplitude-thereby overcoming the inherent limitations of their classical counterparts. We validate this principle on both synthetic high-frequency functions and quantum-advantage tasks. The results show that QNNs significantly outperform DNNs in high-frequency tasks, offering an explanation for QNNs' superior expressivity in complex spectral landscapes.
Leveraging the principles of quantum mechanics—specifically superposition and parallelism— quantum computation holds the potential to surpass the computational power limitations of classical computers. However, due to constraints such as the limited number of available qubits and the presence of decoherence noise in quantum hardware, designing efficient quantum algorithms capable of running on Noisy Intermediate-Scale Quantum (NISQ) devices has become a pivotal research focus. The Variational Quantum Algorithm (VQA) represents a hybrid quantum-classical computational paradigm that is executable on NISQ devices and can potentially demonstrate quantum advantage. In VQAs, the loss function is evaluated on a quantum computer, optimized by a classical computer, and the updated parameters are subsequently fed back to the quantum circuit for the next iteration. The Quantum Convolutional Neural Network (QCNN) is a prominent implementation within the VQA framework. It performs convolutional and pooling operations via Parameterized Quantum Circuits (PQCs) , endowing it with potentially superior feature extraction capabilities compared to its classical convolutional neural network counterparts. This paper begins with the foundational theories of quantum computing, provides a detailed exposition of the VQA workflow and QCNN architecture, and summarizes VQAs based on PQCs. It then reviews recent research progress on QCNNs from a practical application standpoint and concludes by outlining current challenges and future research directions.
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 convolutional neural networks (QCNNs) are emerging as promising models in quantum machine learning, particularly for image classification and computer vision tasks. Recent developments include hybrid classical–quantum architectures, advanced quantum encoding methods, and novel circuit designs that improve data processing on Noisy Intermediate-Scale Quantum (NISQ) devices. However, practical implementation remains challenging due to circuit complexity, gate count, qubit connectivity, and hardware noise, which limit scalability and performance. Consequently, quantum circuit optimization has become essential for reducing resource requirements and improving classification accuracy. This study presents a systematic literature review of 40 research papers published between 2014 and 2025. The review covers QCNNs together with closely related quantum neural network (QNN) models and quantum circuit optimization studies, since circuit-optimization techniques are frequently developed for QNNs more broadly rather than for QCNN architectures in isolation. Within this scope, it examines network architectures, encoding strategies, application domains, and optimization techniques, with particular attention to heuristic and metaheuristic approaches such as genetic algorithms and evolutionary strategies. The findings highlight growing trends in hybrid quantum–classical integration, the widespread adoption of metaheuristic optimization, and the importance of multi-objective frameworks adapted to quantum hardware constraints. Finally, the review identifies key research gaps and future directions for practical QCNN deployment on near-term quantum devices.
Quantum Neural Networks (QNNs) represent a novel computational paradigm that merges the principles of quantum computing with the architecture of artificial neural networks. Through the quantum phenomena of superposition, entanglement, and interference, QNNs enable parallel computation in high-dimensional Hilbert spaces, offering the potential to surpass the representational limits of classical models. This review provides a comprehensive overview of the theoretical foundations and architectures of QNNs, including Quantum Perceptrons, Variational Quantum Circuits (VQCs), Quantum Convolutional Neural Networks (QCNNs), and Quantum Recurrent Neural Networks (QRNNs). Furthermore, it discusses hybrid quantum–classical training mechanisms and key challenges such as barren plateaus, decoherence, and sampling complexity. The review also highlights recent applications of QNNs in medical diagnostics, materials science, and financial forecasting, demonstrating their potential to accelerate computation and improve predictive accuracy. Finally, future research directions are discussed in relation to computational efficiency, model interpretability, and integration with next-generation quantum hardware.
Quantum Neural Networks (QNNs) represent a compelling convergence of quantum mechanics and machine learning, yet their practical deployment remains constrained by noise, limited qubit counts, and the cost of full quantum execution. Hybrid quantum-classical (HQC) training offers a pragmatic path forward: parameterized quantum circuits (PQCs) process information in exponentially large Hilbert spaces while classical optimizers update parameters via gradient methods accessible on today's hardware. We present a modular HQC training framework, formalising the parameter-shift rule for hardware-compatible gradient estimation, characterising the barren-plateau phenomenon through a variance-decay analysis, and introducing noise-aware training strategies for NISQ devices. Empirical results from state-vector and shot-based simulation confirm measurable quantum advantage on quantum-structured tasks, with the phase classification benchmark improving +12.1% over a classical MLP baseline. The accompanying open-source library provides a clean abstraction between quantum simulation and classical optimisation, serving as a foundation for near-term quantum machine learning research.