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

arXiv:2604.06007 · International Joint Conference on Neural Networks (IJCNN)

Scaling Laws for Hybrid Quantum Neural Networks: Depth, Width, and Quantum-Centric Diagnostics

Danil Vyskubov, Kirill Vyskubov, Nouhaila Innan, Muhammad Shafique

On Aqora:Muhammad ShafiqueMuhammad Shafique
quant-phSubmitted Apr 7, 2026• Announced Apr 8, 2026• 9 citations
arXivHTMLPDF

Abstract

Hybrid quantum neural networks are increasingly explored for classification, yet it remains unclear how their performance and quantum behavior scale with circuit depth and qubit count. We present a controlled scaling study of hybrid quantum-classical classifiers along two axes: (1) increasing the number of quantum layers L at fixed qubits Q, and (2) increasing the number of qubits Q at fixed depth L. Across multiple datasets, we evaluate predictive performance using Accuracy, PR-AUC, Precision, Recall, and F1, and track quantum-specific metrics (QCE, EEE, QGN) to characterize how quantum properties evolve under scaling. Our results summarize scaling trends, saturation regimes, and dataset-dependent sensitivity, and further analyze how quantum metrics relate to predictive performance. This study provides practical guidance for selecting (Q,L) in hybrid QNN classifiers and establishes a consistent evaluation protocol for scaling analysis.

Journal reference
2026 International Joint Conference on Neural Networks (IJCNN), Maastricht, Netherlands, 2026, pp. 1-8
DOI
10.1109/IJCNN69854.2026.11729919
Comments
8 pages, 10 figures. Accepted at IJCNN 2026

Related papers

Can the Quantum Fisher Effective Dimension Judge a Hybrid Quantum Neural Network?

Md Masum Mia, Amith Khandakar, Mohamed Arselene Ayari

Hybrid quantum neural networks (QNNs) pair a classical feature extractor with a variational quantum circuit, and the effective dimension of the circuit's quantum Fisher information matrix (QFIM) is widely used to assess such models. This scalar is often interpreted as an indicator of trainability or accuracy, although these readings are rarely tested under controlled conditions. We therefore investigate what it actually measures. We define the effective dimension as the participation ratio of the QFIM diagonal, which lies between one and the parameter count, and we evaluate it across two image datasets, four frozen backbones including a random-projection control, four circuit depths on eight qubits, and three seeds. During training, the effective dimension remains well below the parameter count, and its correlation with test accuracy ranges from -0.72 to +0.87, changing sign across seeds in half of the configurations. Increasing depth activates a larger share of the circuit directions, from 0.38 to 0.91 of the parameter count, while the initialization gradient variance remains between 1e-5 and 1e-4, so no barren plateau (BP) appears at eight qubits. At a fixed circuit, the four backbones share an initial effective dimension of 6 to 9, even though CIFAR-10 accuracy ranges from 23.14% to 86.44%. These results show that the effective dimension characterizes circuit geometry rather than model quality. We conclude with a practical protocol that assigns each diagnostic to the question it can answer.

quant-pharXiv:2610.14070•Oct 9, 2026•0 citations

Scalability Analysis of Quantum Machine Learning Models on Noisy Intermediate-Scale Quantum Devices

S. Afrin Banu, P Jose

Often portrayed as a promising application domain for noisy intermediate-scale quantum (NISQ) computing, quantum machine learning (QML) suffers from limited scalability due to circuit depth and qubit quality, measurement overhead, optimization instability, and noise accumulation. The purpose of the current research paper is to conduct a simulation-based scalability analysis of five representative QML models, namely, the Variational Quantum Classifier, Quantum Support Vector Machine, Quantum Neural Network, Quantum Convolutional Neural Network, and Hybrid Quantum Classical Neural Network. The Breast Cancer Wisconsin benchmark dataset is utilized in this study, and dimensionality reduction is performed to enable qubit-efficient encoding. Furthermore, the effect of increased qubit count, circuit depth, parameter count, and depolarizing noise is examined on model behaviour. The parameters for assessing performance are accuracy, precision, recall, F1 measure, training time, inference time, simulation success probability, noise robustness, and a unified scalability measure. Statistical treatment includes descriptive statistics, one-way ANOVA, Kruskal-Wallis test, correlation analysis and multiple linear regression. The outcomes suggest that the Hybrid Quantum-Classical Neural Network boasted the best overall scalability profile, with the highest mean accuracy and the strongest resistance to noise, while deeper QNN circuits showed heavier degradation at higher levels of noise. According to the results, performance of QML is a near-term one. Similarly, this performance is not determined only by representation, expressivity but other factors as well. The study states that the experimental advantage of QML on NISQ devices is still problem dependent and constrained to simple problems. Shallow hybrid architecture, trivial quantum kernel, error mitigation and hardware adaptive ansatz construction are the realistic pathways for future quantum artificial intelligence systems.

International Conference on Intelligent Cyber Physical Systems and Internet of Things (ICoICI)•Sep 1, 2026•0 citations

Hybrid Quantum-Classical Neural Networks on High-Dimensional Imbalanced Tabular Data: A Convergence Efficiency Analysis

Aditya Shintre, Aditya Wankhede, Sonali Ajankar

Hybrid Quantum-Classical Neural Networks (HQNNs) offer a practical approach for near-term quantum machine learning; however, their performance on real-world, high-dimensional, and imbalanced tabular datasets remains underexplored. In this work, we benchmark HQNNs against parameter-equivalent classical models on two data-driven classification tasks: the Santander Customer Transaction dataset (200 features, 9:1 class imbalance) and the UCI Adult dataset (105 features, 3:1 class imbalance). Across 48 independent experimental trials, we evaluate 24 hybrid quantum configurations alongside 24 classical baselines by varying qubit count (N ∈ {4,6}), circuit depth (L ∈ {3,4,5}), and entanglement topology (Strongly Entangling Layers vs. Random Layers). Results show that select hybrid configurations reduce training epochs by a factor of 2–4 while maintaining competitive AUC; for example, on Santander, peak AUC is reached at 15 epochs compared to 60 for classical models, with only a 0.0166 difference in final AUC. We further observe that structured entanglement yields more stable training, and while deeper (5-layer) configurations can exhibit training variance, successful runs still yield strong convergence efficiencies. These findings highlight HQNNs as an efficient and practical approach for data-driven imbalanced classification tasks.

IEEE International Conference on Instrumentation (INSTCon)•Jul 1, 2026•0 citations

Hybrid Quantum-Classical Neural Networks: Bridging Deep Learning and Quantum Computing for Image Classification

Akash Jha, Rahul Bhandari, Raman Chadha

Given the recent prominence of Noisy Intermediate Scale Quantum (NISQ) devices, hybrid quantum-classical neural networks have recently drawn a lot of excitement as a potential solution to improving learning efficiency. Our current paper describes a Hybrid Quantum-Classical Convolutional Neural Network (HQCNN) which is defined as a pretrained backbone of ResNet-18 and a small quantum-enhanced feature-transformation layer of Qiskit. The convolutional high-dimensional features are reduced to the classical eight-dimensional one, which in turn is encoded into a four-qubit parameterized quantum circuit. The resultant quantum expectation values are combined with the compressed features to come up with a hybridized representation of twelve dimensions, used to classify multi-class dataset. Experimental analysis of CIFAR10 data has shown that the suggested HQCNN achieves a test accuracy of 94.74% and improves class-wise precision, recall, and F1-scores than a number of existing hybrid models while using a significantly smaller number of latent features. More investigation into the quantum module indicates that it is predictable about scalability to batch, resistant to perturbations in features, and efficiently performed by evaluating circuit vectors. These results indicate that compact and expressive representations of images that can be helpful in the NISQ era can be acquired using hybrid architectures, thus providing a practical way to forward the benefits of quantum to modern deep-learning pipelines.

International Conference on Emerging Computational Intelligence (ICECI)•Apr 4, 2026•0 citations

A multi-axis diagnostic framework for Hybrid Quantum-Classical Neural Networks in Empirical setting

Praneel Gore

Abstract Hybrid quantum-classical neural networks (HQNNs) are a promising paradigm in quantum enhaced machine learning, yet their evaluation remains an open challenge constrained by two key issues: reliance on single metrics (e.g., loss or gradients) and synthetic benchmarks that fail to reflect real-world complexities. This work introduces a multi-metric diagnostic framework, that offers a holistic view of HQNN behavior by analyzing primary axes: quantum trainability (gradient norms), model convergence (loss curves), and quantum utility (via the Quantum Contribution Score, QCS). And secondary axes: expressivity, gradient efficiency and entanglement entropy. Across 16 models over real-world and synthetic data with a classical artificial neural network as baseline. Through this framework empirical phenomena in practical evaluations are developed, including gradient recovery after barren plateaus, data and expressivity dependent QCS trends, relation between gradient behaviour and qubit to layer ratio and model convergence despite barren plateaus.

•Jul 17, 2025•0 citations

Depth-dependent separability growth in variational quantum neural networks and its empirical coupling to a parameter-sensitivity statistic

Zhihai Xie, Shibin Zhang, Mian Ren

Abstract Understanding how depth-dependent quantum evolution reshapes the geometric organization of data representations in quantum neural networks (QNNs) remains a central question in quantum machine learning. We study a hardware-efficient variational QNN—single-qubit R Y / R Z rotations per layer followed by a nearest-neighbor CNOT entangling ring, with circuit depth n ranging from 3 to 12 layers—and ask how its data representation geometry reorganizes as n increases. We report two findings across MNIST and BloodMNIST (15 independent random seeds for amplitude encoding: 5 MNIST, 10 BloodMNIST), with a BloodMNIST angle-encoding control (10 further seed-runs, 25 in total) used to check the second finding’s recurrence under a different encoding strategy. First, class separability shows power-law growth within the tested depth window, S ( n ) ∝ n b , where b = δ − ε captures the balance between inter-class expansion and intra-class diffusion; this description holds throughout the tested range for MNIST, whereas for BloodMNIST the growth is better described by a model approaching a depth-independent ceiling within the same range, with b retained as a common cross-dataset summary exponent. Second, we measure an auxiliary parameter-perturbation sensitivity statistic and find that its depth-trend exponent γ shows a dataset-level association with b via b ≈ k data ( 1 − | γ | ) . We explicitly tested whether this statistic tracks the textbook state-based quantum Fisher information by computing the latter rigorously for the same trained circuits, and found that it does not; we also find that the b – γ association is weak and not statistically significant within either dataset alone. We therefore report this coupling descriptively, as an open dataset-level pattern rather than an established physical law. To our knowledge, this is the first report of this depth-dependent separability-growth pattern in QNNs together with a practical, predictive heuristic for depth selection derived from it, using each dataset’s own functional form. In the tested depth regime ( n ∈ [ 3 , 12 ]

Physica Scripta•Jul 9, 2026•0 citations
Hybrid quantum neural networks are increasingly explored for classification, yet it remains unclear how their performance and quantum behavior scale with circuit depth and qubit count. We present a controlled scaling study of hybrid quantum-classical classifiers along two axes: (1) increasing the number of quantum layers L at fixed qubits Q, and (2) increasing the number of qubits Q at fixed depth L. Across multiple datasets, we evaluate predictive performance using Accuracy, PR-AUC, Precision, Recall, and F1, and track quantum-specific metrics (QCE, EEE, QGN) to characterize how quantum properties evolve under scaling. Our results summarize scaling trends, saturation regimes, and dataset-dependent sensitivity, and further analyze how quantum metrics relate to predictive performance. This study provides practical guidance for selecting (Q,L) in hybrid QNN classifiers and establishes a consistent evaluation protocol for scaling analysis.
Hybrid quantum neural networks (QNNs) pair a classical feature extractor with a variational quantum circuit, and the effective dimension of the circuit's quantum Fisher information matrix (QFIM) is widely used to assess such models. This scalar is often interpreted as an indicator of trainability or accuracy, although these readings are rarely tested under controlled conditions. We therefore investigate what it actually measures. We define the effective dimension as the participation ratio of the QFIM diagonal, which lies between one and the parameter count, and we evaluate it across two image datasets, four frozen backbones including a random-projection control, four circuit depths on eight qubits, and three seeds. During training, the effective dimension remains well below the parameter count, and its correlation with test accuracy ranges from -0.72 to +0.87, changing sign across seeds in half of the configurations. Increasing depth activates a larger share of the circuit directions, from 0.38 to 0.91 of the parameter count, while the initialization gradient variance remains between 1e-5 and 1e-4, so no barren plateau (BP) appears at eight qubits. At a fixed circuit, the four backbones share an initial effective dimension of 6 to 9, even though CIFAR-10 accuracy ranges from 23.14% to 86.44%. These results show that the effective dimension characterizes circuit geometry rather than model quality. We conclude with a practical protocol that assigns each diagnostic to the question it can answer.
Often portrayed as a promising application domain for noisy intermediate-scale quantum (NISQ) computing, quantum machine learning (QML) suffers from limited scalability due to circuit depth and qubit quality, measurement overhead, optimization instability, and noise accumulation. The purpose of the current research paper is to conduct a simulation-based scalability analysis of five representative QML models, namely, the Variational Quantum Classifier, Quantum Support Vector Machine, Quantum Neural Network, Quantum Convolutional Neural Network, and Hybrid Quantum Classical Neural Network. The Breast Cancer Wisconsin benchmark dataset is utilized in this study, and dimensionality reduction is performed to enable qubit-efficient encoding. Furthermore, the effect of increased qubit count, circuit depth, parameter count, and depolarizing noise is examined on model behaviour. The parameters for assessing performance are accuracy, precision, recall, F1 measure, training time, inference time, simulation success probability, noise robustness, and a unified scalability measure. Statistical treatment includes descriptive statistics, one-way ANOVA, Kruskal-Wallis test, correlation analysis and multiple linear regression. The outcomes suggest that the Hybrid Quantum-Classical Neural Network boasted the best overall scalability profile, with the highest mean accuracy and the strongest resistance to noise, while deeper QNN circuits showed heavier degradation at higher levels of noise. According to the results, performance of QML is a near-term one. Similarly, this performance is not determined only by representation, expressivity but other factors as well. The study states that the experimental advantage of QML on NISQ devices is still problem dependent and constrained to simple problems. Shallow hybrid architecture, trivial quantum kernel, error mitigation and hardware adaptive ansatz construction are the realistic pathways for future quantum artificial intelligence systems.
Hybrid Quantum-Classical Neural Networks (HQNNs) offer a practical approach for near-term quantum machine learning; however, their performance on real-world, high-dimensional, and imbalanced tabular datasets remains underexplored. In this work, we benchmark HQNNs against parameter-equivalent classical models on two data-driven classification tasks: the Santander Customer Transaction dataset (200 features, 9:1 class imbalance) and the UCI Adult dataset (105 features, 3:1 class imbalance). Across 48 independent experimental trials, we evaluate 24 hybrid quantum configurations alongside 24 classical baselines by varying qubit count (N ∈ {4,6}), circuit depth (L ∈ {3,4,5}), and entanglement topology (Strongly Entangling Layers vs. Random Layers). Results show that select hybrid configurations reduce training epochs by a factor of 2–4 while maintaining competitive AUC; for example, on Santander, peak AUC is reached at 15 epochs compared to 60 for classical models, with only a 0.0166 difference in final AUC. We further observe that structured entanglement yields more stable training, and while deeper (5-layer) configurations can exhibit training variance, successful runs still yield strong convergence efficiencies. These findings highlight HQNNs as an efficient and practical approach for data-driven imbalanced classification tasks.
Given the recent prominence of Noisy Intermediate Scale Quantum (NISQ) devices, hybrid quantum-classical neural networks have recently drawn a lot of excitement as a potential solution to improving learning efficiency. Our current paper describes a Hybrid Quantum-Classical Convolutional Neural Network (HQCNN) which is defined as a pretrained backbone of ResNet-18 and a small quantum-enhanced feature-transformation layer of Qiskit. The convolutional high-dimensional features are reduced to the classical eight-dimensional one, which in turn is encoded into a four-qubit parameterized quantum circuit. The resultant quantum expectation values are combined with the compressed features to come up with a hybridized representation of twelve dimensions, used to classify multi-class dataset. Experimental analysis of CIFAR10 data has shown that the suggested HQCNN achieves a test accuracy of 94.74% and improves class-wise precision, recall, and F1-scores than a number of existing hybrid models while using a significantly smaller number of latent features. More investigation into the quantum module indicates that it is predictable about scalability to batch, resistant to perturbations in features, and efficiently performed by evaluating circuit vectors. These results indicate that compact and expressive representations of images that can be helpful in the NISQ era can be acquired using hybrid architectures, thus providing a practical way to forward the benefits of quantum to modern deep-learning pipelines.
Abstract Hybrid quantum-classical neural networks (HQNNs) are a promising paradigm in quantum enhaced machine learning, yet their evaluation remains an open challenge constrained by two key issues: reliance on single metrics (e.g., loss or gradients) and synthetic benchmarks that fail to reflect real-world complexities. This work introduces a multi-metric diagnostic framework, that offers a holistic view of HQNN behavior by analyzing primary axes: quantum trainability (gradient norms), model convergence (loss curves), and quantum utility (via the Quantum Contribution Score, QCS). And secondary axes: expressivity, gradient efficiency and entanglement entropy. Across 16 models over real-world and synthetic data with a classical artificial neural network as baseline. Through this framework empirical phenomena in practical evaluations are developed, including gradient recovery after barren plateaus, data and expressivity dependent QCS trends, relation between gradient behaviour and qubit to layer ratio and model convergence despite barren plateaus.
Abstract Understanding how depth-dependent quantum evolution reshapes the geometric organization of data representations in quantum neural networks (QNNs) remains a central question in quantum machine learning. We study a hardware-efficient variational QNN—single-qubit R Y / R Z rotations per layer followed by a nearest-neighbor CNOT entangling ring, with circuit depth n ranging from 3 to 12 layers—and ask how its data representation geometry reorganizes as n increases. We report two findings across MNIST and BloodMNIST (15 independent random seeds for amplitude encoding: 5 MNIST, 10 BloodMNIST), with a BloodMNIST angle-encoding control (10 further seed-runs, 25 in total) used to check the second finding’s recurrence under a different encoding strategy. First, class separability shows power-law growth within the tested depth window, S ( n ) ∝ n b , where b = δ − ε captures the balance between inter-class expansion and intra-class diffusion; this description holds throughout the tested range for MNIST, whereas for BloodMNIST the growth is better described by a model approaching a depth-independent ceiling within the same range, with b retained as a common cross-dataset summary exponent. Second, we measure an auxiliary parameter-perturbation sensitivity statistic and find that its depth-trend exponent γ shows a dataset-level association with b via b ≈ k data ( 1 − | γ | ) . We explicitly tested whether this statistic tracks the textbook state-based quantum Fisher information by computing the latter rigorously for the same trained circuits, and found that it does not; we also find that the b – γ association is weak and not statistically significant within either dataset alone. We therefore report this coupling descriptively, as an open dataset-level pattern rather than an established physical law. To our knowledge, this is the first report of this depth-dependent separability-growth pattern in QNNs together with a practical, predictive heuristic for depth selection derived from it, using each dataset’s own functional form. In the tested depth regime ( n ∈ [ 3 , 12 ]