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:2310.04120 · Advanced Quantum Technologies

A General Approach to Dropout in Quantum Neural Networks

Francesco Scala, Andrea Ceschini, Massimo Panella, Dario Gerace

Published Dec 10, 2023• 25 citations
arXivHTMLPDF

Abstract

Abstract In classical machine learning (ML), “overfitting” is the phenomenon occurring when a given model learns the training data excessively well, and it thus performs poorly on unseen data. A commonly employed technique in ML is the so called “dropout,” which prevents computational units from becoming too specialized, hence reducing the risk of overfitting. With the advent of quantum neural networks (QNNs) as learning models, overfitting might soon become an issue, owing to the increasing depth of quantum circuits as well as multiple embedding of classical features, which are employed to give the computational nonlinearity. Here, a generalized approach is presented to apply the dropout technique in QNN models, defining and analyzing different quantum dropout strategies to avoid overfitting and achieve a high level of generalization. This study allows to envision the power of quantum dropout in enabling generalization, providing useful guidelines on determining the maximal dropout probability for a given model, based on overparametrization theory. It also highlights how quantum dropout does not impact the features of the QNN models, such as expressibility and entanglement. All these conclusions are supported by extensive numerical simulations and may pave the way to efficiently employing deep quantum machine learning (QML) models based on state‐of‐the‐art QNNs.

DOI
10.1002/qute.202300220

Related papers

Post-training approach for mitigating overfitting in quantum convolutional neural networks

Aakash Ravindra Shinde, Charu Jain, Amir Kalev

Quantum convolutional neural network (QCNN), an early application for quantum computers in the noisy intermediate-scale quantum era, has been consistently proven successful as a machine learning (ML) algorithm for several tasks with significant accuracy. Derived from its classical counterpart, QCNN is prone to overfitting. Overfitting is a typical shortcoming of ML models that are trained too closely to the availed training dataset and perform relatively poorly on unseen datasets for a similar problem. In this work we study post-training approaches for mitigating overfitting in QCNNs. We find that a straightforward adaptation of a classical post-training method, known as neuron dropout, to the quantum setting leads to a significant and undesirable consequence: a substantial decrease in success probability of the QCNN. We argue that this effect exposes the crucial role of entanglement in QCNNs and the vulnerability of QCNNs to entanglement loss. Hence, we propose a parameter adaptation method as an alternative method. Our method is computationally efficient and is found to successfully handle overfitting in the test cases.

arXiv:2309.01829•Physical Review A•Oct 11, 2024•6 citations

Softer is Better: Tweaking Quantum Dropout to Enhance Quantum Neural Network Trainability

Daniele Lizzio Bosco, Riccardo Romanello, Giuseppe Serra +1 more

Quantum Machine Learning (QML) has been acknowledged for its transformative potential in enhancing computational capabilities beyond classical approaches, with Quantum Neural Networks (QNNs) emerging as one of the most promising models. Despite the advancements, the adaptation of effective optimization techniques such as batch normalization or regularization, which are well-established in classical Neural Networks, remains an ongoing challenge for QNNs.This adaptation is crucial, especially when considering the implementation of techniques like quantum dropout, which, although inspired by its classical counterpart to reduce the risk of overfitting by "deleting" certain model components during the forward step, exhibits distinct behaviours in a quantum context. This study introduces a novel approach to modulate the impact of dropout in QNNs through a technique we term Soft Dropout. Our method introduces a parameterized softening mechanism for gate eliminations, enabling a more nuanced control over the dropout process, thereby mitigating its adverse effects on the network’s learning capacity, and reducing at the same time the risk of overfitting.Our experimental analysis demonstrates that softening current quantum dropout consistently enhances model performance across a spectrum of configurations. This improvement is attributed to the intrinsic properties of quantum gates, which allow for a gradual adjustment of their impact on the quantum circuit up to the identity operation. The results highlight the importance of adapting classical machine learning techniques to the quantum context, considering the unique computational model of quantum processing, potentially accelerating the adoption of QML in real-world problems.

International Conference on Quantum Communications, Networking, and Computing (QCNC)•Mar 31, 2025•0 citations

Generalization despite overfitting in quantum machine learning models

Evan Peters, Maria Schuld

The widespread success of deep neural networks has revealed a surprise in classical machine learning: very complex models often generalize well while simultaneously overfitting training data. This phenomenon of benign overfitting has been studied for a variety of classical models with the goal of better understanding the mechanisms behind deep learning. Characterizing the phenomenon in the context of quantum machine learning might similarly improve our understanding of the relationship between overfitting, overparameterization, and generalization. In this work, we provide a characterization of benign overfitting in quantum models. To do this, we derive the behavior of a classical interpolating Fourier features models for regression on noisy signals, and show how a class of quantum models exhibits analogous features, thereby linking the structure of quantum circuits (such as data-encoding and state preparation operations) to overparameterization and overfitting in quantum models. We intuitively explain these features according to the ability of the quantum model to interpolate noisy data with locally "spiky" behavior and provide a concrete demonstration example of benign overfitting.

arXiv:2209.05523•Quantum•Dec 20, 2023•99 citations

Quantum Machine Learning: A Step towards the Hybrid Quantum Classical Approaches

Kailash Thakur, Parveen Sadotra, Shaina Mahajan +2 more

Machine Learning (ML) has changed the way computers work by letting them learn from data and make smart choices. However, traditional ML techniques encounter scalability and efficiency constraints as datasets get larger and computational problems become more complicated[1]. These restrictions could be circumvented by the new paradigms that quantum computing offers by using entanglement and superposition [2], [3]. The combination of the Quantum computing with the Machine Learning is called the Quantum Machine Learning(QML), offers efficiency improvements, new algorithmic opportunities, and speedups in certain linear algebraic operations. So, this review dives deeply into quantum-classical algorithms, with a particular emphasis on the approaches like quantum neural network and quantum support vector machines. We examine computational complexity, architectural design, benchmarking findings, and real-world applications (such as optimization, cryptography, and classification). We also discuss important issues including scalability, data encoding, and noise, and we suggest future paths for error prevention and scalable structures.

IEEE International Conference on Recent Trends in Computing and Smart Mobility (RCSM)•Dec 5, 2025•0 citations

Quantum Neural Networks

Syed Naseer Ahmad Shah, Sabia Hassan, Kaiser Javeed Giri

Quantum neural networks (QNNs) are a revolutionary paradigm placed at the frontier of machine learning and quantum computing. By utilizing quantum mechanical principles like superposition, entanglement, and quantum parallelism, QNNs facilitate the solution of complex machine learning problems that were thought to be impossible according to conventional models. This chapter provides a detailed study of QNNs, from elementary principles to practical applications. It begins with a rundown of the most significant concepts in machine learning and quantum computing, followed by QNN architecture, with important points including quantum gates and parameterized quantum circuits. The chapter addresses challenges in training QNNs, with special attention given to hybrid quantum-classical optimization algorithms and their utility in successful model training. It also examines multiple applications of QNNs, including simulation in quantum chemistry, optimization problems, and real-world machine learning applications, to demonstrate how they have the potential to outperform conventional algorithms in high dimensional or highly entangled data environments. This chapter provides a comparative assessment of quantum machine learning frameworks, quantum hardware platforms, and hybrid optimization methods and aids in determining the most suitable tools to design QNNs. It finally discusses possible areas of future research and research issues in the field, emphasizing the urgent need to improve quantum hardware, increase noise tolerance, and scale up. As quantum computing technology evolves, QNNs are expected to be at the center of solving some of the most complex and computationally intensive problems encountered in various industries.

Auerbach Publications eBooks•Feb 24, 2026•0 citations

Q-Drop: Optimizing Quantum Orthogonal Networks with Statistic Pruning and Dynamic Dropout

Pham Thai Quang Nguyen, Tran Cat Khanh, Yared Abera Ergu +1 more

Quantum machine learning (QML) holds immense potential for revolutionizing computational intelligence, yet faces significant challenges in optimizing quantum neural network architectures for practical implementation. This paper introduces Q-Drop, an innovative quantum adaptation technique that enhances parameterized quantum circuits (PQCs) for training performance. The system includes two novel optimization strategies: statistical pruning and dynamic dropout (circuit flipflops), which significantly improve the learning performance and robustness of quantum orthogonal neural networks. Evaluation results across standard (MNIST, Fashion-MNIST) and medical imaging (Pneumonia-MNIST, Retina-MNIST) datasets demonstrate substantial performance gains. Notably, the approach achieves up to 94.3 % accuracy on medical image classification and 98.7 % accuracy on standard image datasets, outperforming existing quantum machine learning methods. By addressing critical challenges in quantum neural network training, this work highlights the potential of applying quantum machine learning to enhance performance for classical computer vision tasks and QML-driven scheduling optimization in quantum networks.

International Conference on Communications•Jun 8, 2025•1 citation
Abstract In classical machine learning (ML), “overfitting” is the phenomenon occurring when a given model learns the training data excessively well, and it thus performs poorly on unseen data. A commonly employed technique in ML is the so called “dropout,” which prevents computational units from becoming too specialized, hence reducing the risk of overfitting. With the advent of quantum neural networks (QNNs) as learning models, overfitting might soon become an issue, owing to the increasing depth of quantum circuits as well as multiple embedding of classical features, which are employed to give the computational nonlinearity. Here, a generalized approach is presented to apply the dropout technique in QNN models, defining and analyzing different quantum dropout strategies to avoid overfitting and achieve a high level of generalization. This study allows to envision the power of quantum dropout in enabling generalization, providing useful guidelines on determining the maximal dropout probability for a given model, based on overparametrization theory. It also highlights how quantum dropout does not impact the features of the QNN models, such as expressibility and entanglement. All these conclusions are supported by extensive numerical simulations and may pave the way to efficiently employing deep quantum machine learning (QML) models based on state‐of‐the‐art QNNs.
Quantum convolutional neural network (QCNN), an early application for quantum computers in the noisy intermediate-scale quantum era, has been consistently proven successful as a machine learning (ML) algorithm for several tasks with significant accuracy. Derived from its classical counterpart, QCNN is prone to overfitting. Overfitting is a typical shortcoming of ML models that are trained too closely to the availed training dataset and perform relatively poorly on unseen datasets for a similar problem. In this work we study post-training approaches for mitigating overfitting in QCNNs. We find that a straightforward adaptation of a classical post-training method, known as neuron dropout, to the quantum setting leads to a significant and undesirable consequence: a substantial decrease in success probability of the QCNN. We argue that this effect exposes the crucial role of entanglement in QCNNs and the vulnerability of QCNNs to entanglement loss. Hence, we propose a parameter adaptation method as an alternative method. Our method is computationally efficient and is found to successfully handle overfitting in the test cases.
Quantum Machine Learning (QML) has been acknowledged for its transformative potential in enhancing computational capabilities beyond classical approaches, with Quantum Neural Networks (QNNs) emerging as one of the most promising models. Despite the advancements, the adaptation of effective optimization techniques such as batch normalization or regularization, which are well-established in classical Neural Networks, remains an ongoing challenge for QNNs.This adaptation is crucial, especially when considering the implementation of techniques like quantum dropout, which, although inspired by its classical counterpart to reduce the risk of overfitting by "deleting" certain model components during the forward step, exhibits distinct behaviours in a quantum context. This study introduces a novel approach to modulate the impact of dropout in QNNs through a technique we term Soft Dropout. Our method introduces a parameterized softening mechanism for gate eliminations, enabling a more nuanced control over the dropout process, thereby mitigating its adverse effects on the network’s learning capacity, and reducing at the same time the risk of overfitting.Our experimental analysis demonstrates that softening current quantum dropout consistently enhances model performance across a spectrum of configurations. This improvement is attributed to the intrinsic properties of quantum gates, which allow for a gradual adjustment of their impact on the quantum circuit up to the identity operation. The results highlight the importance of adapting classical machine learning techniques to the quantum context, considering the unique computational model of quantum processing, potentially accelerating the adoption of QML in real-world problems.
The widespread success of deep neural networks has revealed a surprise in classical machine learning: very complex models often generalize well while simultaneously overfitting training data. This phenomenon of benign overfitting has been studied for a variety of classical models with the goal of better understanding the mechanisms behind deep learning. Characterizing the phenomenon in the context of quantum machine learning might similarly improve our understanding of the relationship between overfitting, overparameterization, and generalization. In this work, we provide a characterization of benign overfitting in quantum models. To do this, we derive the behavior of a classical interpolating Fourier features models for regression on noisy signals, and show how a class of quantum models exhibits analogous features, thereby linking the structure of quantum circuits (such as data-encoding and state preparation operations) to overparameterization and overfitting in quantum models. We intuitively explain these features according to the ability of the quantum model to interpolate noisy data with locally "spiky" behavior and provide a concrete demonstration example of benign overfitting.
Machine Learning (ML) has changed the way computers work by letting them learn from data and make smart choices. However, traditional ML techniques encounter scalability and efficiency constraints as datasets get larger and computational problems become more complicated[1]. These restrictions could be circumvented by the new paradigms that quantum computing offers by using entanglement and superposition [2], [3]. The combination of the Quantum computing with the Machine Learning is called the Quantum Machine Learning(QML), offers efficiency improvements, new algorithmic opportunities, and speedups in certain linear algebraic operations. So, this review dives deeply into quantum-classical algorithms, with a particular emphasis on the approaches like quantum neural network and quantum support vector machines. We examine computational complexity, architectural design, benchmarking findings, and real-world applications (such as optimization, cryptography, and classification). We also discuss important issues including scalability, data encoding, and noise, and we suggest future paths for error prevention and scalable structures.
Quantum neural networks (QNNs) are a revolutionary paradigm placed at the frontier of machine learning and quantum computing. By utilizing quantum mechanical principles like superposition, entanglement, and quantum parallelism, QNNs facilitate the solution of complex machine learning problems that were thought to be impossible according to conventional models. This chapter provides a detailed study of QNNs, from elementary principles to practical applications. It begins with a rundown of the most significant concepts in machine learning and quantum computing, followed by QNN architecture, with important points including quantum gates and parameterized quantum circuits. The chapter addresses challenges in training QNNs, with special attention given to hybrid quantum-classical optimization algorithms and their utility in successful model training. It also examines multiple applications of QNNs, including simulation in quantum chemistry, optimization problems, and real-world machine learning applications, to demonstrate how they have the potential to outperform conventional algorithms in high dimensional or highly entangled data environments. This chapter provides a comparative assessment of quantum machine learning frameworks, quantum hardware platforms, and hybrid optimization methods and aids in determining the most suitable tools to design QNNs. It finally discusses possible areas of future research and research issues in the field, emphasizing the urgent need to improve quantum hardware, increase noise tolerance, and scale up. As quantum computing technology evolves, QNNs are expected to be at the center of solving some of the most complex and computationally intensive problems encountered in various industries.
Quantum machine learning (QML) holds immense potential for revolutionizing computational intelligence, yet faces significant challenges in optimizing quantum neural network architectures for practical implementation. This paper introduces Q-Drop, an innovative quantum adaptation technique that enhances parameterized quantum circuits (PQCs) for training performance. The system includes two novel optimization strategies: statistical pruning and dynamic dropout (circuit flipflops), which significantly improve the learning performance and robustness of quantum orthogonal neural networks. Evaluation results across standard (MNIST, Fashion-MNIST) and medical imaging (Pneumonia-MNIST, Retina-MNIST) datasets demonstrate substantial performance gains. Notably, the approach achieves up to 94.3 % accuracy on medical image classification and 98.7 % accuracy on standard image datasets, outperforming existing quantum machine learning methods. By addressing critical challenges in quantum neural network training, this work highlights the potential of applying quantum machine learning to enhance performance for classical computer vision tasks and QML-driven scheduling optimization in quantum networks.