Jul 2026· 2026 International Conference on Intelligent and Sustainable AI Systems (ICOSAAS)· pp. 727-732· 0 citations· 16 references
Abstract
Quantum machine learning (QML) faces practical limitations due to noisy intermediate-scale quantum (NISQ) constraints, including noise, restricted qubit availability, and unstable optimization. This paper proposes HyQNet, a resource-aware hybrid quantum–classical framework designed to address these challenges through efficient circuit execution and adaptive optimization. The framework integrates optimized quantum circuits with classical learning strategies to improve scalability and stability under NISQ conditions. Experimental results on Iris, Wine, and Breast Cancer datasets show that HyQNet achieves an accuracy of 95.1% and F1-score of 94.8%, outperforming variational QNN (92.6%) and quantum SVM (91.2%). It also reduces runtime to 16.9 s compared to 20.5 s for VQNN, while maintaining efficient utilization of 8 qubits. Statistical analysis confirms significance (p < 0.05), and ablation studies validate the contribution of each component. The results demonstrate improved convergence stability and resource efficiency in hybrid quantum learning systems.
Optimization problems are critical across various domains, yet existing quantum algorithms, despite their great potential, struggle with scalability and accuracy due to excessive reliance on entanglement. To address these limitations, we propose a variational quantum optimization algorithm (VQOA), which employs an ansatz based solely on quantum superposition, where single-qubit rotation gates function analogously to neurons in classical deep neural networks. This ansatz, which can be regarded as quantum neural networks, significantly reduces circuit complexity, enhances noise robustness, mitigates Barren Plateau issues, and enables efficient partitioning for highly complex, large-scale optimization. Furthermore, we introduce distributed VQOA (DVQOA), which integrates high-performance computing with quantum computing to achieve superior performance. These features enable a significant acceleration of material optimization tasks (e.g., metamaterial design), achieving more than 50 × speedup compared to state-of-the-art optimization algorithms. Beyond material design, DVQOA efficiently solves quantum chemistry problems and
N
-ary (
N
≥2) optimization problems involving higher-order interactions, outperforming classical deep neural networks. These advantages establish DVQOA as a highly promising and versatile solver for real-world problems, demonstrating the practical benefits of the quantum-classical approach.
Seongmin Kim, In-Saeng Suh· npj Computational Materials· 7 citations
This paper examines quantum algorithms and their computational complexity through a unified framework combining mathematical modeling, system architecture, and empirical evaluation. Key complexity measures circuit depth, gate count, and query complexity are analyzed under NISQ constraints. A hybrid quantum classical optimization framework is introduced to improve efficiency and stability. Results show the full model achieves 94.2% accuracy with baseline runtime, while removing optimization lowers accuracy to 85.6% and increases runtime by 30%. Reducing qubits decreases cost but drops accuracy to 78.3%, and disabling error mitigation causes unstable performance at 70.1%. Comparative analysis indicates strong advantages of quantum algorithms for structured problems, especially in scalability and asymptotic complexity. However, performance remains sensitive to noise, limited qubits, and circuit depth, emphasizing the need for hardware–algorithm co-design.
Quantum Machine Learning (QML) has emerged as a promising interdisciplinary field that combines quantum computing with machine learning to address complex computational problems. This survey provides a comprehensive overview of the theoretical foundations, unified taxonomy, and key methodologies in QML. We discuss quantum data encoding techniques, variational quantum models, neural-inspired quantum architectures, and quantum data learning approaches, along with their associated challenges and limitations. The survey further examines evaluation strategies, the concept of quantum advantage, and the role of current quantum hardware and software platforms in advancing QML research. In addition, major application domains and the growing importance of trustworthy QML, including interpretability, robustness, and security, are explored. Finally, we highlight open challenges and future research directions to provide insights into the evolving landscape of Quantum Machine Learning.
Ashis Kumar Pati, Rajesh Vayyala, K. Mohanty et al.· IEEE Access· 0 citations
Next-generation HQCNN architectures, including DMERA, HEA–QCNN, and (bQCNN) designs, significantly optimized for NISQ hardware are presented, establishing a scalable, parameter-efficient framework for high-performance hybrid quantum-classical ML on near-term devices.
Shyam R. Sihare, A. Cherukuri· SN Computer Science· 0 citations
This work analyzes how circuit design constraints can systematically reduce the measurement overhead associated with repeated evaluations of the candidate gate pool in adaptive algorithms by focusing on the Hadamard test circuit architecture, hardware-aware qubit connectivity, and problem-specific adaptive framework.
This study systematically evaluates the efficacy of two quantum machine learning algorithms-Quantum Support Vector Machine (QSVM) and Quantum Neural Networks (QNN) on IBM quantum simulation platforms to provide a comprehensive assessment of QSVM and QNN under current Noisy Intermediate-Scale Quantum (NISQ) constraints.