Quantum Deep Learning (QDL) is in its infancy and is a potential domain where classical and quantum technology intersect well to speed up many classical algorithms to reduce time complexity from exponential to polynomial. Thus, this systematic review article aims to look deeper into this domain and show the common architectures, building blocks, and integration approaches, as well as challenges and opportunities for computer scientists. Our research shows that the three main ways QDL is implemented are through quantum-inspired, hybrid classical-quantum, and fully fledged quantum algorithms. For now, the limitations of quantum hardware have led to the dominance of hybrid classical-quantum over fully fledged quantum and quantum-inspired due to the issue with noise in deep quantum circuits. In QDL, Variational Quantum Algorithms (VQAs) utilizing Parameterized Quantum Circuits (PQCs) act as the heart of the neural networks, much like tensors do in classical deep learning algorithms. This quantum circuit is also known as an ansatz and can be executed iteratively using quantum computation, with the circuit parameters optimized using either classical or quantum optimizers. The integrated ansatzes are shown to solve problems more efficiently than classical algorithms and open more opportunities for computer scientists interested in research in a domain historically dominated by physics.
Quantum Deep Learning (QDL) is in its infancy and is a potential domain where classical and quantum technology intersect well to speed up many classical algorithms to reduce time complexity from exponential to polynomial. Thus, this systematic review article aims to look deeper into this domain and show the common architectures, building blocks, and integration approaches, as well as challenges and opportunities for computer scientists. Our research shows that the three main ways QDL is implemented are through quantum-inspired, hybrid classical-quantum, and fully fledged quantum algorithms. For now, the limitations of quantum hardware have led to the dominance of hybrid classical-quantum over fully fledged quantum and quantum-inspired due to the issue with noise in deep quantum circuits. In QDL, Variational Quantum Algorithms (VQAs) utilizing Parameterized Quantum Circuits (PQCs) act as the heart of the neural networks, much like tensors do in classical deep learning algorithms. This quantum circuit is also known as an ansatz and can be executed iteratively using quantum computation, with the circuit parameters optimized using either classical or quantum optimizers. The integrated ansatzes are shown to solve problems more efficiently than classical algorithms and open more opportunities for computer scientists interested in research in a domain historically dominated by physics.
Quantum Machine Learning (QML) and Quantum Deep Learning (QDL) are emerging research areas that combine quantum computing with machine learning and deep learning techniques.These approaches aim to improve computational speed, feature representation, optimisation, and learning performance in complex data-driven problems.This review paper presents a structured discussion of recent developments in QML and QDL, focusing on major algorithms, application areas, implementation approaches, current limitations, and future research directions.The review covers important quantum-based learning methods, including quantum support vector machines, quantum k-nearest neighbours, quantum principal component analysis, quantum neural networks, variational quantum classifiers, quantum convolutional neural networks, quantum reinforcement learning, quantum generative models, and hybrid quantum-classical models.The paper also discusses how these methods are being explored in healthcare, finance, cybersecurity, communication networks, image processing, natural language processing, and optimisation.However, the review shows that most QML and QDL models are still in the early stages of research.Current challenges include noisy quantum hardware, limited qubit count, data encoding issues, circuit design problems, difficulties with model training, and a lack of standardised benchmarks.The review concludes that QML and QDL have strong future potential, but their practical success depends on improved hardware, stronger algorithms, reliable frameworks, and fair comparison with classical models.
Quantum deep learning (QDL) explores the use of both quantum and quantum-inspired resources to determine when deep learning's core capabilities, such as expressivity, generalization, and scalability, can be enhanced based on specific resource constraints. Distinct from broader quantum machine learning, QDL emphasizes compositional depth at the pipeline level and the integration of quantum or quantum-inspired components within end-to-end workflows. This review provides an operational definition of QDL and introduces a taxonomy comprising four primary paradigms: hybrid quantum-classical models, quantum deep neural networks, quantum algorithms for deep learning primitives, and quantum-inspired classical algorithms. Theoretical principles are connected to advanced architectures, software toolchains, and experimental demonstrations across superconducting, trapped-ion, photonic, semiconductor spin, and neutral-atom systems, as well as quantum annealers. Claims of quantum advantage are critically assessed by distinguishing provable complexity-theoretic separations from empirical observations. The analysis characterizes trade-offs between model expressivity, trainability, and classical simulability, while systematically detailing the bottlenecks imposed by optimization landscapes, input-output access models, and hardware constraints. Applications are surveyed in domains encompassing image classification, natural language processing, scientific discovery, quantum data processing, and quantum optimal control, underscoring fair benchmarking against optimized classical counterparts and a comprehensive assessment of resource requirements. This review serves as a tutorial entry point for graduate students while guiding readers to specialized literature. It concludes with a verification-aware roadmap to transition QDL from near-term demonstrations to scalable and fault-tolerant implementations.
Quantum deep learning (QDL) for image classification has emerged as a rapidly growing research area at the intersection of quantum computing and computer vision. However, the literature remains fragmented across diverse model architectures, datasets, and evaluation settings, making it difficult to obtain a focused understanding of the field. This study presents a systematic review of QDL models for image classification published between 2020 and 2025. The review was conducted using a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-based selection process supported by a mapping perspective. Relevant studies were identified from major academic databases, screened using predefined inclusion and exclusion criteria, and analyzed through a taxonomy-based framework. A total of 47 studies were included and organized into four main architectural categories: Quantum Neural Networks (QNNs), Quantum Convolutional Neural Networks (QCNNs), Quantum Vision Transformers (QViT), and Quantum Attention Mechanisms. The findings show that most existing approaches rely on hybrid quantum-classical designs, where quantum components are integrated selectively into specific stages of the learning pipeline rather than replacing the full classical architecture. Several studies report encouraging results, such as improved parameter efficiency, competitive ACC (accuracy), and robustness in specific settings. Current progress remains constrained by hardware noise, limited qubit availability, encoding overhead, scalability challenges, and inconsistent benchmarking practices. It remains a promising but still emerging direction, with future progress expected to depend on more scalable architectures, fairer evaluation protocols, and broader validation on realistic application.
Quantum deep learning (QDL) explores the use of both quantum and quantum-inspired resources to determine when deep learning's core capabilities, such as expressivity, generalization, and scalability, can be enhanced based on specific resource constraints. Distinct from broader quantum machine learning, QDL emphasizes compositional depth at the pipeline level and the integration of quantum or quantum-inspired components within end-to-end workflows. This review provides an operational definition of QDL and introduces a taxonomy comprising four primary paradigms: hybrid quantum-classical models, quantum deep neural networks, quantum algorithms for deep learning primitives, and quantum-inspired classical algorithms. Theoretical principles are connected to advanced architectures, software toolchains, and experimental demonstrations across superconducting, trapped-ion, photonic, semiconductor spin, and neutral-atom systems, as well as quantum annealers. Claims of quantum advantage are critically assessed by distinguishing provable complexity-theoretic separations from empirical observations. The analysis characterizes trade-offs between model expressivity, trainability, and classical simulability, while systematically detailing the bottlenecks imposed by optimization landscapes, input-output access models, and hardware constraints. Applications are surveyed in domains encompassing image classification, natural language processing, scientific discovery, quantum data processing, and quantum optimal control, underscoring fair benchmarking against optimized classical counterparts and a comprehensive assessment of resource requirements. This review serves as a tutorial entry point for graduate students while guiding readers to specialized literature. It concludes with a verification-aware roadmap to transition QDL from near-term demonstrations to scalable and fault-tolerant implementations.
Hybrid Quantum Deep Learning (HQDL) models have emerged as a pragmatic bridge between the algorithmic richness of deep learning and the computational advantages of near-term quantum devices. Rather than seeking full quantum replacement of neural architectures, current research favours layered or modular integration-where variational quantum circuits, quantum kernels, or quantum-enhanced feature embeddings operate alongside classical neural networks. This convergence has yielded promising results in domains such as classification, generative modelling, molecular simulation, and combinatorial optimization. However, performance varies significantly depending on data encoding strategies, circuit expressivity, and training stability under hardware noise. This review synthesizes the taxonomy of existing HQDL architectures, evaluates their empirical benefits over purely classical baselines, and contextualizes their feasibility within current Noisy Intermediate-Scale Quantum (NISQ) constraints. Furthermore, it highlights reproducibility challenges, benchmarking inconsistencies, and open questions surrounding optimization landscapes and scalability. The analysis concludes that while HQDL remains in an early experimental phase, its trajectory is rapidly transitioning from proof-of-concept demonstrations toward targeted domain specialization in finance, materials discovery, and strategic decision-making.
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.