Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE Transactions on Mobile Computing

Efficient Quantum Neural Architecture Search With Filtered Supernet for Mobile Learning Systems

Seok Bin Son, Hyojun Ahn, Samuel Yen-Chi Chen, Soohyun Park, Joongheon Kim

Published Jul 1, 2026• 0 citations
Publisher

Abstract

In low-connectivity mobile environments, image processing must be performed on-device due to limited access to central servers. However, deep learning models specialized for image processing demand heavy computation, rendering them unsuitable in these environments. To overcome these limitations, quantum neural networks (QNNs) provide a lightweight alternative, achieving high performance with fewer parameters by leveraging quantum advantage. QNN performance relies on the parameterized quantum circuit (PQC) structure, but identifying optimal designs is challenging due to the vast search space and hardware constraints. Applying the same QNN across heterogeneous mobile devices often results in degraded performance under varying constraints. Neural architecture search (NAS) offers an automated approach to optimizing PQC structures for device-specific limitations, although conventional methods are costly and frequently overlook quantum constraints. This paper introduces GreedyQNAS, a lightweight quantum NAS framework inspired by GreedyNAS and GreedyNASv2. It improves search efficiency by eliminating weak architectures during training, defined as low-performing candidates. GreedyQNAS adopts a two-stage search strategy: amacro level searchusing deep reinforcement learning for gate placement location under device-level constraints, and amicro level searchusing filtered one-shot NAS to select optimal gate types. On the medical-MNIST dataset, the method reduces parameters by 77% while preserving high accuracy, confirming its efficiency for mobile quantum simulation inference.

DOI
10.1109/tmc.2026.3709164

Related papers

Q-RLONAS: Towards Efficient Quantum Neural Architecture Search

Seok Bin Son, Soohyun Park

Quantum neural architecture search aims to automate the design of parameterized quantum circuits (PQCs) within quantum neural networks (QNNs), where learning performance is highly sensitive to the placement and type of quantum gates. However, applying traditional NAS techniques to quantum circuits presents significant challenges, including barren plateaus, noise accumulation, device-specific constraints, and the high computational cost of independently evaluating all candidate architectures. To address these limitations, this paper proposes Q-RLONAS, a two-stage quantum NAS framework that integrates deep reinforcement learning (DRL) with one-shot NAS to efficiently explore and optimize PQC structures. In the Stage #1, DRL determines the optimal gate placement by leveraging a reward function that considers performance, noise, barren plateaus, and hardware constraints. In the Stage #2, a one-shot NAS method is employed to select effective quantum gate types using weight sharing within a supernet, significantly reducing training overhead. Experimental results on the mini-MNIST dataset demonstrate that Q-RLONAS achieves high accuracy while satisfying gate placement constraints, outperforming existing approaches in terms of efficiency for quantum circuit design.

IEEE International Conference on Quantum Computing and Engineering (QCE)•Aug 30, 2025•2 citations

Quantum-Inspired Neural Architecture Search (Q-NAS)

Adithya Jakkaraju

Neural Architecture Search (NAS) has revolutionized deep learning by automating the design of optimal neural network architectures. However, classical NAS methods suffer from high computational costs and inefficient search processes. In this research, we introduce Quantum-Inspired Neural Architecture Search (Q-NAS), leveraging quantum computing paradigms such as quantum annealing and variational quantum circuits (VQC) to optimize NAS for edge-device deployment. We explore hybrid quantum-classical frameworks that improve search efficiency and reduce energy consumption. Our work provides comparative analyses, proposes novel quantum search strategies, and benchmarks Q-NAS against traditional NAS methods. Experimental results demonstrate significant reductions in search time and computational costs while maintaining or improving model accuracy.

Journal of Electrical Systems•Jan 25, 2024•0 citations

Differentiable Hardware-Aware Quantum Neural Architecture Search for Resource-Constrained Medical Imaging

Sanvi Sharma, Panchadip Bhattacharjee, Somyajeet Arukh +3 more

Quantum machine learning (QML) provides parameter-efficient substitutes for traditional deep networks in medical imaging settings with constrained computational resources. However, deployment on noisy intermediate-scale quantum (NISQ) devices is difficult due to manual architecture design and decoherence constraints, particularly when circuit depth is limited. We introduce D-HAQNAS (Differentiable HardwareAware Quantum Neural Architecture Search), a differentiable hardware-aware framework that uses end-to-end differentiable optimization to find energy-efficient quantum topologies. Our method outperforms evolutionary methods in architecture search by up to$15 \times$. Three essential elements are integrated into the suggested framework: (1) a differentiable super-circuit with sigmoid-gated architectural parameters that permits gradient-based single-qubit gate topology search; (2) hardware-aware penalty terms that constrain gate count and circuit depth during training; and (3) weighted residual fusion to stabilize hybrid classical-quantum gradient flow. D-HAQNAS achieves 96.21% accuracy on PneumoniaMNIST, which is statistically comparable to classical baselines but with 78.6% fewer trainable parameters and a 31.5% decrease in the number of gates (from 54 to 37). D-HAQNAS shows statistically significant improvements on the severely data-limited RetinaMNIST dataset (1,080 samples, 5-class diabetic retinopathy classification) with an accuracy of 57.50% compared to 53.33% for classical models (+4.17 percentage point absolute gain, paired$t$-test$p=0.031)$. These findings suggest that parameter-efficient hybrid quantum-classical networks are a viable strategy for clinical settings with limited data, especially those where parameter efficiency is crucial.

IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT)•Jul 1, 2026•0 citations

QNAS: A Neural Architecture Search Framework for Accurate and Efficient Quantum Neural Networks

Kooshan Maleki, Alberto Marchisio, Muhammad Shafique

Designing quantum neural networks (QNNs) that are both accurate and deployable on NISQ hardware is challenging. Handcrafted ansatze must balance expressivity, trainability, and resource use, while limited qubits often necessitate circuit cutting. Existing quantum architecture search methods primarily optimize accuracy while only heuristically controlling quantum and mostly ignore the exponential overhead of circuit cutting. We introduce QNAS, a neural architecture search framework that unifies hardware aware evaluation, multi objective optimization, and cutting overhead awareness for hybrid quantum classical neural networks (HQNNs). QNAS trains a shared parameter SuperCircuit and uses NSGA-II to optimize three objectives jointly: (i) validation error, (ii) a runtime cost proxy measuring wall clock evaluation time, and (iii) the estimated number of subcircuits under a target qubit budget. QNAS evaluates candidate HQNNs under a few epochs of training and discovers clear Pareto fronts that reveal tradeoffs between accuracy, efficiency, and cutting overhead. Across MNIST, Fashion-MNIST, and Iris benchmarks, we observe that embedding type and CNOT mode selection significantly impact both accuracy and efficiency, with angle-y embedding and sparse entangling patterns outperforming other configurations on image datasets, and amplitude embedding excelling on tabular data (Iris). On MNIST, the best architecture achieves 97.16% test accuracy with a compact 8 qubit, 2 layer circuit; on the more challenging Fashion-MNIST, 87.38% with a 5 qubit, 2 layer circuit; and on Iris, 100% validation accuracy with a 4 qubit, 2 layer circuit. QNAS surfaces these design insights automatically during search, guiding practitioners toward architectures that balance accuracy, resource efficiency, and practical deployability on current hardware.

quant-pharXiv:2604.07013•International Joint Conference on Neural Networks (IJCNN)•Apr 9, 2026•4 citations

Quantum Circuit Structure Optimization for Quantum Reinforcement Learning

Seok Bin Son, Joongheon Kim

Reinforcement learning (RL) enables agents to learn optimal policies through environmental interaction. However, RL suffers from reduced learning efficiency due to the curse of dimensionality in high-dimensional spaces. Quantum reinforcement learning (QRL) addresses this issue by leveraging superposition and entanglement in quantum computing, allowing efficient handling of high-dimensional problems with fewer resources. QRL combines quantum neural networks (QNNs) with RL, where the parameterized quantum circuit (PQC) acts as the core computational module. The PQC performs linear and nonlinear transformations through gate operations, similar to hidden layers in classical neural networks. Previous QRL studies, however, have used fixed PQC structures based on empirical intuition without verifying their optimality. This paper proposes a QRL-NAS algorithm that integrates quantum neural architecture search (QNAS) to optimize PQC structures within QRL. Experiments demonstrate that QRL-NAS achieves higher rewards than QRL with fixed circuits, validating its effectiveness and practical utility.

arXiv:2507.00589•Jul 1, 2025•0 citations

Neural Architecture Search for Quantum Autoencoders

Hibah Agha, Samuel Yen-Chi Chen, Huan-Hsin Tseng +1 more

In recent years, machine learning and deep learning have driven advances in domains such as image classification, speech recognition, and anomaly detection by leveraging multi-layer neural networks to model complex data. Simultaneously, quantum computing (QC) promises to address classically intractable problems via quantum parallelism, motivating research in quantum machine learning (QML). Among QML techniques, quantum autoencoders show promise for compressing high-dimensional quantum and classical data. However, designing effective quantum circuit architectures for quantum autoencoders remains challenging due to the complexity of selecting gates, arranging circuit layers, and tuning parameters. This paper proposes a neural architecture search (NAS) framework that automates the design of quantum autoencoders using a genetic algorithm (GA). By systematically evolving variational quantum circuit (VQC) configurations, our method seeks to identify high-performing hybrid quantum-classical autoencoders for data reconstruction without becoming trapped in local minima. We demonstrate effectiveness on image datasets, highlighting the potential of quantum autoencoders for efficient feature extraction within a noise-prone, near-term quantum era. Our approach lays a foundation for broader application of genetic algorithms to quantum architecture search, aiming for a robust, automated method that can adapt to varied data and hardware constraints.

quant-pharXiv:2511.19246•Nov 25, 2025•0 citations
In low-connectivity mobile environments, image processing must be performed on-device due to limited access to central servers. However, deep learning models specialized for image processing demand heavy computation, rendering them unsuitable in these environments. To overcome these limitations, quantum neural networks (QNNs) provide a lightweight alternative, achieving high performance with fewer parameters by leveraging quantum advantage. QNN performance relies on the parameterized quantum circuit (PQC) structure, but identifying optimal designs is challenging due to the vast search space and hardware constraints. Applying the same QNN across heterogeneous mobile devices often results in degraded performance under varying constraints. Neural architecture search (NAS) offers an automated approach to optimizing PQC structures for device-specific limitations, although conventional methods are costly and frequently overlook quantum constraints. This paper introduces GreedyQNAS, a lightweight quantum NAS framework inspired by GreedyNAS and GreedyNASv2. It improves search efficiency by eliminating weak architectures during training, defined as low-performing candidates. GreedyQNAS adopts a two-stage search strategy: amacro level searchusing deep reinforcement learning for gate placement location under device-level constraints, and amicro level searchusing filtered one-shot NAS to select optimal gate types. On the medical-MNIST dataset, the method reduces parameters by 77% while preserving high accuracy, confirming its efficiency for mobile quantum simulation inference.
Quantum neural architecture search aims to automate the design of parameterized quantum circuits (PQCs) within quantum neural networks (QNNs), where learning performance is highly sensitive to the placement and type of quantum gates. However, applying traditional NAS techniques to quantum circuits presents significant challenges, including barren plateaus, noise accumulation, device-specific constraints, and the high computational cost of independently evaluating all candidate architectures. To address these limitations, this paper proposes Q-RLONAS, a two-stage quantum NAS framework that integrates deep reinforcement learning (DRL) with one-shot NAS to efficiently explore and optimize PQC structures. In the Stage #1, DRL determines the optimal gate placement by leveraging a reward function that considers performance, noise, barren plateaus, and hardware constraints. In the Stage #2, a one-shot NAS method is employed to select effective quantum gate types using weight sharing within a supernet, significantly reducing training overhead. Experimental results on the mini-MNIST dataset demonstrate that Q-RLONAS achieves high accuracy while satisfying gate placement constraints, outperforming existing approaches in terms of efficiency for quantum circuit design.
Neural Architecture Search (NAS) has revolutionized deep learning by automating the design of optimal neural network architectures. However, classical NAS methods suffer from high computational costs and inefficient search processes. In this research, we introduce Quantum-Inspired Neural Architecture Search (Q-NAS), leveraging quantum computing paradigms such as quantum annealing and variational quantum circuits (VQC) to optimize NAS for edge-device deployment. We explore hybrid quantum-classical frameworks that improve search efficiency and reduce energy consumption. Our work provides comparative analyses, proposes novel quantum search strategies, and benchmarks Q-NAS against traditional NAS methods. Experimental results demonstrate significant reductions in search time and computational costs while maintaining or improving model accuracy.
Quantum machine learning (QML) provides parameter-efficient substitutes for traditional deep networks in medical imaging settings with constrained computational resources. However, deployment on noisy intermediate-scale quantum (NISQ) devices is difficult due to manual architecture design and decoherence constraints, particularly when circuit depth is limited. We introduce D-HAQNAS (Differentiable HardwareAware Quantum Neural Architecture Search), a differentiable hardware-aware framework that uses end-to-end differentiable optimization to find energy-efficient quantum topologies. Our method outperforms evolutionary methods in architecture search by up to15×15 \times15×. Three essential elements are integrated into the suggested framework: (1) a differentiable super-circuit with sigmoid-gated architectural parameters that permits gradient-based single-qubit gate topology search; (2) hardware-aware penalty terms that constrain gate count and circuit depth during training; and (3) weighted residual fusion to stabilize hybrid classical-quantum gradient flow. D-HAQNAS achieves 96.21% accuracy on PneumoniaMNIST, which is statistically comparable to classical baselines but with 78.6% fewer trainable parameters and a 31.5% decrease in the number of gates (from 54 to 37). D-HAQNAS shows statistically significant improvements on the severely data-limited RetinaMNIST dataset (1,080 samples, 5-class diabetic retinopathy classification) with an accuracy of 57.50% compared to 53.33% for classical models (+4.17 percentage point absolute gain, pairedttt-testp=0.031)p=0.031)p=0.031). These findings suggest that parameter-efficient hybrid quantum-classical networks are a viable strategy for clinical settings with limited data, especially those where parameter efficiency is crucial.
Designing quantum neural networks (QNNs) that are both accurate and deployable on NISQ hardware is challenging. Handcrafted ansatze must balance expressivity, trainability, and resource use, while limited qubits often necessitate circuit cutting. Existing quantum architecture search methods primarily optimize accuracy while only heuristically controlling quantum and mostly ignore the exponential overhead of circuit cutting. We introduce QNAS, a neural architecture search framework that unifies hardware aware evaluation, multi objective optimization, and cutting overhead awareness for hybrid quantum classical neural networks (HQNNs). QNAS trains a shared parameter SuperCircuit and uses NSGA-II to optimize three objectives jointly: (i) validation error, (ii) a runtime cost proxy measuring wall clock evaluation time, and (iii) the estimated number of subcircuits under a target qubit budget. QNAS evaluates candidate HQNNs under a few epochs of training and discovers clear Pareto fronts that reveal tradeoffs between accuracy, efficiency, and cutting overhead. Across MNIST, Fashion-MNIST, and Iris benchmarks, we observe that embedding type and CNOT mode selection significantly impact both accuracy and efficiency, with angle-y embedding and sparse entangling patterns outperforming other configurations on image datasets, and amplitude embedding excelling on tabular data (Iris). On MNIST, the best architecture achieves 97.16% test accuracy with a compact 8 qubit, 2 layer circuit; on the more challenging Fashion-MNIST, 87.38% with a 5 qubit, 2 layer circuit; and on Iris, 100% validation accuracy with a 4 qubit, 2 layer circuit. QNAS surfaces these design insights automatically during search, guiding practitioners toward architectures that balance accuracy, resource efficiency, and practical deployability on current hardware.
Reinforcement learning (RL) enables agents to learn optimal policies through environmental interaction. However, RL suffers from reduced learning efficiency due to the curse of dimensionality in high-dimensional spaces. Quantum reinforcement learning (QRL) addresses this issue by leveraging superposition and entanglement in quantum computing, allowing efficient handling of high-dimensional problems with fewer resources. QRL combines quantum neural networks (QNNs) with RL, where the parameterized quantum circuit (PQC) acts as the core computational module. The PQC performs linear and nonlinear transformations through gate operations, similar to hidden layers in classical neural networks. Previous QRL studies, however, have used fixed PQC structures based on empirical intuition without verifying their optimality. This paper proposes a QRL-NAS algorithm that integrates quantum neural architecture search (QNAS) to optimize PQC structures within QRL. Experiments demonstrate that QRL-NAS achieves higher rewards than QRL with fixed circuits, validating its effectiveness and practical utility.
In recent years, machine learning and deep learning have driven advances in domains such as image classification, speech recognition, and anomaly detection by leveraging multi-layer neural networks to model complex data. Simultaneously, quantum computing (QC) promises to address classically intractable problems via quantum parallelism, motivating research in quantum machine learning (QML). Among QML techniques, quantum autoencoders show promise for compressing high-dimensional quantum and classical data. However, designing effective quantum circuit architectures for quantum autoencoders remains challenging due to the complexity of selecting gates, arranging circuit layers, and tuning parameters. This paper proposes a neural architecture search (NAS) framework that automates the design of quantum autoencoders using a genetic algorithm (GA). By systematically evolving variational quantum circuit (VQC) configurations, our method seeks to identify high-performing hybrid quantum-classical autoencoders for data reconstruction without becoming trapped in local minima. We demonstrate effectiveness on image datasets, highlighting the potential of quantum autoencoders for efficient feature extraction within a noise-prone, near-term quantum era. Our approach lays a foundation for broader application of genetic algorithms to quantum architecture search, aiming for a robust, automated method that can adapt to varied data and hardware constraints.