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

IEIE Transactions on Smart Processing and Computing

Research on the Design of Financial Data Analysis Platform with Joint PSO-HQCNN and RPA Robot Visualization Technology

Zhenzhen Tang

Published Oct 31, 2024• 0 citations
Publisher

Abstract

A novel RPA technique for financial data analysis based on quantum neural network model is innovatively developed. Compared to the conventional model, the Hybrid Quantum Convolutional Neural Network model (HQCNN) reduces the loss value by 0.0056% and improves the training and testing accuracy by 0.93% and 0.36%, respectively. In addition, the hybrid quantum convolutional neural network model of this RPA exhibits less fluctuation in loss value after 60 iterations and remains stable after 90 iterations. Meanwhile, we combine the particle swarm optimization algorithm with HQCNN to transform classical data into quantum state data for joint optimization. The results of the recognition and validation accuracy plot show that PSOHQCNN has the best recognition effect, with a recognition accuracy of 96.85%, followed by HQCNN with a recognition accuracy of 96.20%. And CNN has the lowest recognition accuracy of 95.65%. However, PSO-HQCNN model increases the time cost while improving the accuracy. Therefore, in the process of applying RPA to financial systems, trade-offs should be made according to the actual situation, and the HQCNN model or PSO-HQCNN model should be selected. In conclusion, through the implementation process of RPA robot visualization application established by using the quantum neural network model, this study provides a new direction of exploration and a feasible implementation path for the automated process processing of enterprise financial data analysis.

DOI
10.5573/ieiespc.2024.10.13.5.472

Related papers

Research on image classification using quantum convolutional neural networks based on multiple approaches

Yumin Dong, Xiaofeng Ni, Tonglei Sun +2 more

Abstract With the significant enhancement of quantum computing capabilities and their expanding applications in various fields, quantum image processing combined with quantum computing has become a key area of research in quantum computing, holding important theoretical value. In the face of the challenges posed by the digital information era, image classification technology demands higher standards from machine learning algorithms in terms of accuracy, computational speed, generalization capability, and complexity. This paper studies a series of quantum data encoding methods and parametric quantum circuit architectures, and proposes the application of Hybrid Hierarchical Encoding (HHE) and Hybrid Quantum Convolutional Neural Network (HQCNN). The new encoding method HHE integrates the benefits of angle encoding and enhances the hybrid amplitude encoding design to achieve optimal classification performance without compromising the depth of quantum circuits.HQCNN demonstrates outstanding classification performance and remarkable control over model complexity by leveraging the data processing capabilities of classical convolution and the complex computational power of quantum convolution. Experimental results show that HHE and HQCNN achieve better classification accuracy on the MNIST and Fashion-MNIST datasets compared to traditional classical CNN models, particularly when the number of parameters is significantly reduced.

Physica Scripta•Mar 18, 2025•2 citations

Application of image recognition algorithm based on quantum convolutional neural network

Xueyang Zhao

In this paper, an innovative Quantum image recognition algorithm, Quantum Auto Gradient Descent (QAGD), is proposed. The algorithm combines the advantages of quantum convolutional neural networks (QCNN), autoencoders (AE) and low-lot gradient descent (MGD). The introduction of an automatic encoder into the quantum convolutional neural network can effectively extract image features and reduce dimensionality, reduce data dimensions while retaining key information, and improve the model's ability to capture image features. At the same time, combined with the small-batch gradient descent optimization strategy, the model parameters are iteratively updated through small-batch data, which speeds up the model convergence speed and improves the training efficiency. In the experimental part, we compare QAGD with traditional image recognition algorithms, and the results show that QAGD algorithm has significant advantages in improving image recognition performance and reducing false positive rate, with higher recognition accuracy and faster training speed. This fully verifies the feasibility and effectiveness of QAGD architecture in the application of image recognition algorithms based on quantum convolutional neural networks.

IEEE International Conference on Electronic Technology, Communication and Information (ICETCI)•May 23, 2025•0 citations

Hybrid Quantum-Classical Neural Network (HQCNN) for Crypto Price Forecasting

Md Shujan Shak, Nabila F. Rahman, Fuad Mahmud +3 more

Cryptocurrency price prediction remains a challenging problem due to the highly volatile nature of digital asset markets. This study introduces a Hybrid Quantum-Classical Neural Network (HQCNN) for cryptocurrency price forecasting, leveraging quantum feature encoding and extraction with deep learning architectures. The proposed model utilizes Quantum Convolutional Neural Networks (QCNN) to enhance feature representations, which are then processed by classical deep learning models such as Long Short-Term Memory (LSTM) and Transformer networks. Experimental results on historical cryptocurrency data demonstrate that HQCNN outperforms state-of-the-art models. The proposed model achieves a mean absolute error (MAE) of 0.023 and a root mean squared error (RMSE) of 0.036, surpassing traditional deep learning methods such as LSTM (MAE: 0.040, RMSE: 0.056) and Transformerbased models (MAE: 0.035, RMSE: 0.050). Ablation studies confirm the importance of quantum feature encoding, as removing it degrades performance by up to $\mathbf{3 8. 2 \%}$. Additionally, HQCNN reduces inference time to 7 milliseconds, making it suitable for real-time trading applications. Despite its advantages, HQCNN faces challenges related to quantum hardware constraints, training overhead, and interpretability in financial decision-making. Future work will focus on deploying the model on real quantum processors, integrating reinforcement learning for market adaptability, and enhancing interpretability with quantum explainability techniques.

International Conference on Artificial Intelligence, Blockchain, and Internet of Things (AIBThings)•Sep 6, 2025•0 citations

Data Augmentation Effectiveness Analysis on Improving QCNN Algorithm Performance

Shidqi Indy Izhari, Infall Syafalni, Nur Ulfa Maulidevi +1 more

Quantum Convolutional Neural Networks (QCNNs) represent an emerging paradigm that combines quantum computing principles with convolutional neural network concepts. However, limitations in dataset size and quantum processing complexity often hinder optimal performance. This research analyzes the effectiveness of various data augmentation techniques in improving QCNN algorithm performance, examining both pure quantum models and hybrid approaches with classical-to-quantum transfer learning. Data augmentation techniques are categorized into geometric transformations, photometric transformations, and their combinations. Experimental results demonstrate that augmentation significantly impacts full QCNN models, particularly when combined with angle encoding, achieving accuracy improvements up to 87.00 % with an average enhancement of 27.27 % compared to baseline without augmentation. For hybrid QCNN models, photometric transformations (P3) with saturation and hue adjustments achieved maximum accuracy of 73.00 % and average improvement of 12.93 % using ResNet18 architecture. These findings reinforce the potential of data augmentation as an important strategy in developing more accurate QCNN models.

International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)•Nov 4, 2025•0 citations

An Optimized Hybrid Quantum-Classical Neural Network Model for Handwritten Digit Classification

Vu-Quoc-Minh Nguyen, Trung-Khanh Le, Trong-Tu Bui +1 more

Quantum computers are the latest dream of humans in solving nondeterministic polynomial problems. These problems include machine learning and object classification. However, commercial quantum computers are still not available. In the meantime, scientists developed hybrid models by combining Convolutional Neural Networks (CNN) and Quantum Neural Networks (QNN) to apply quantum computing to artificial intelligence (AI) applications. The hybrid model accelerates the processing of an AI machine by combining the advantage of CNN in feature extraction with the parallel computing capability of QNN. In this study, we optimized the existing hybrid models with a fewer number of qubits to improve the performance of QNN on digit recognition. The study also presents the implementation of the model using Qiskit, describes the training process, and analyzes the impact of the number of qubits and quantum gates on the classification results. Experiments show that Hybrid Quantum-Classical Neural Networks (H-QNN) have the potential to outperform pure CNN. Finally, the study suggests future improvements, including the optimization of quantum algorithms and extending experiments to more complex data sets.

International Conference on Integrated Circuits, Design, and Verification (ICDV)•Jun 16, 2025•0 citations

Optimizing Quantum Neural Networks: A Comparative Study of QNN Architectures and Training Techniques

Lipantri Mashur Gultom, Desi Amirullah

This study explores various architectures of Quantum Neural Networks (QNN) that utilize quantum computing to enhance machine learning model performance. The architectures examined include Hybrid Quantum Neural Network (H-QNN), Deep Quantum Neural Network (DQNN), Quantum Convolutional Neural Network (QCNN), and Coherent Feed Forward Quantum Neural Network. Additionally, the EfficientSU2 architecture demonstrated superior performance in accuracy and cost reduction. Optimization techniques such as Adaptive Moment Estimation (ADAM), Analytic Quantum Gradient Descent (AQGD), and Nakanishi-Fujii-Todo (NFT) were integrated to improve training efficiency. The study results show that EfficientSU2 with 10,000 iterations yielded the best results in terms of cost and accuracy across all tested datasets, with accuracy improvements reaching up to 97%. In conclusion, the EfficientSU2 architecture and the applied optimization techniques significantly enhanced the processing of complex data, paving the way for substantial advancements in quantum-based artificial intelligence applications.

International Conference on Industrial Technology and Business Proceeding (INTECHBIZ)•Dec 24, 2024•0 citations
A novel RPA technique for financial data analysis based on quantum neural network model is innovatively developed. Compared to the conventional model, the Hybrid Quantum Convolutional Neural Network model (HQCNN) reduces the loss value by 0.0056% and improves the training and testing accuracy by 0.93% and 0.36%, respectively. In addition, the hybrid quantum convolutional neural network model of this RPA exhibits less fluctuation in loss value after 60 iterations and remains stable after 90 iterations. Meanwhile, we combine the particle swarm optimization algorithm with HQCNN to transform classical data into quantum state data for joint optimization. The results of the recognition and validation accuracy plot show that PSOHQCNN has the best recognition effect, with a recognition accuracy of 96.85%, followed by HQCNN with a recognition accuracy of 96.20%. And CNN has the lowest recognition accuracy of 95.65%. However, PSO-HQCNN model increases the time cost while improving the accuracy. Therefore, in the process of applying RPA to financial systems, trade-offs should be made according to the actual situation, and the HQCNN model or PSO-HQCNN model should be selected. In conclusion, through the implementation process of RPA robot visualization application established by using the quantum neural network model, this study provides a new direction of exploration and a feasible implementation path for the automated process processing of enterprise financial data analysis.
Abstract With the significant enhancement of quantum computing capabilities and their expanding applications in various fields, quantum image processing combined with quantum computing has become a key area of research in quantum computing, holding important theoretical value. In the face of the challenges posed by the digital information era, image classification technology demands higher standards from machine learning algorithms in terms of accuracy, computational speed, generalization capability, and complexity. This paper studies a series of quantum data encoding methods and parametric quantum circuit architectures, and proposes the application of Hybrid Hierarchical Encoding (HHE) and Hybrid Quantum Convolutional Neural Network (HQCNN). The new encoding method HHE integrates the benefits of angle encoding and enhances the hybrid amplitude encoding design to achieve optimal classification performance without compromising the depth of quantum circuits.HQCNN demonstrates outstanding classification performance and remarkable control over model complexity by leveraging the data processing capabilities of classical convolution and the complex computational power of quantum convolution. Experimental results show that HHE and HQCNN achieve better classification accuracy on the MNIST and Fashion-MNIST datasets compared to traditional classical CNN models, particularly when the number of parameters is significantly reduced.
In this paper, an innovative Quantum image recognition algorithm, Quantum Auto Gradient Descent (QAGD), is proposed. The algorithm combines the advantages of quantum convolutional neural networks (QCNN), autoencoders (AE) and low-lot gradient descent (MGD). The introduction of an automatic encoder into the quantum convolutional neural network can effectively extract image features and reduce dimensionality, reduce data dimensions while retaining key information, and improve the model's ability to capture image features. At the same time, combined with the small-batch gradient descent optimization strategy, the model parameters are iteratively updated through small-batch data, which speeds up the model convergence speed and improves the training efficiency. In the experimental part, we compare QAGD with traditional image recognition algorithms, and the results show that QAGD algorithm has significant advantages in improving image recognition performance and reducing false positive rate, with higher recognition accuracy and faster training speed. This fully verifies the feasibility and effectiveness of QAGD architecture in the application of image recognition algorithms based on quantum convolutional neural networks.
Cryptocurrency price prediction remains a challenging problem due to the highly volatile nature of digital asset markets. This study introduces a Hybrid Quantum-Classical Neural Network (HQCNN) for cryptocurrency price forecasting, leveraging quantum feature encoding and extraction with deep learning architectures. The proposed model utilizes Quantum Convolutional Neural Networks (QCNN) to enhance feature representations, which are then processed by classical deep learning models such as Long Short-Term Memory (LSTM) and Transformer networks. Experimental results on historical cryptocurrency data demonstrate that HQCNN outperforms state-of-the-art models. The proposed model achieves a mean absolute error (MAE) of 0.023 and a root mean squared error (RMSE) of 0.036, surpassing traditional deep learning methods such as LSTM (MAE: 0.040, RMSE: 0.056) and Transformerbased models (MAE: 0.035, RMSE: 0.050). Ablation studies confirm the importance of quantum feature encoding, as removing it degrades performance by up to 38.2%\mathbf{3 8. 2 \%}38.2%. Additionally, HQCNN reduces inference time to 7 milliseconds, making it suitable for real-time trading applications. Despite its advantages, HQCNN faces challenges related to quantum hardware constraints, training overhead, and interpretability in financial decision-making. Future work will focus on deploying the model on real quantum processors, integrating reinforcement learning for market adaptability, and enhancing interpretability with quantum explainability techniques.
Quantum Convolutional Neural Networks (QCNNs) represent an emerging paradigm that combines quantum computing principles with convolutional neural network concepts. However, limitations in dataset size and quantum processing complexity often hinder optimal performance. This research analyzes the effectiveness of various data augmentation techniques in improving QCNN algorithm performance, examining both pure quantum models and hybrid approaches with classical-to-quantum transfer learning. Data augmentation techniques are categorized into geometric transformations, photometric transformations, and their combinations. Experimental results demonstrate that augmentation significantly impacts full QCNN models, particularly when combined with angle encoding, achieving accuracy improvements up to 87.00 % with an average enhancement of 27.27 % compared to baseline without augmentation. For hybrid QCNN models, photometric transformations (P3) with saturation and hue adjustments achieved maximum accuracy of 73.00 % and average improvement of 12.93 % using ResNet18 architecture. These findings reinforce the potential of data augmentation as an important strategy in developing more accurate QCNN models.
Quantum computers are the latest dream of humans in solving nondeterministic polynomial problems. These problems include machine learning and object classification. However, commercial quantum computers are still not available. In the meantime, scientists developed hybrid models by combining Convolutional Neural Networks (CNN) and Quantum Neural Networks (QNN) to apply quantum computing to artificial intelligence (AI) applications. The hybrid model accelerates the processing of an AI machine by combining the advantage of CNN in feature extraction with the parallel computing capability of QNN. In this study, we optimized the existing hybrid models with a fewer number of qubits to improve the performance of QNN on digit recognition. The study also presents the implementation of the model using Qiskit, describes the training process, and analyzes the impact of the number of qubits and quantum gates on the classification results. Experiments show that Hybrid Quantum-Classical Neural Networks (H-QNN) have the potential to outperform pure CNN. Finally, the study suggests future improvements, including the optimization of quantum algorithms and extending experiments to more complex data sets.
This study explores various architectures of Quantum Neural Networks (QNN) that utilize quantum computing to enhance machine learning model performance. The architectures examined include Hybrid Quantum Neural Network (H-QNN), Deep Quantum Neural Network (DQNN), Quantum Convolutional Neural Network (QCNN), and Coherent Feed Forward Quantum Neural Network. Additionally, the EfficientSU2 architecture demonstrated superior performance in accuracy and cost reduction. Optimization techniques such as Adaptive Moment Estimation (ADAM), Analytic Quantum Gradient Descent (AQGD), and Nakanishi-Fujii-Todo (NFT) were integrated to improve training efficiency. The study results show that EfficientSU2 with 10,000 iterations yielded the best results in terms of cost and accuracy across all tested datasets, with accuracy improvements reaching up to 97%. In conclusion, the EfficientSU2 architecture and the applied optimization techniques significantly enhanced the processing of complex data, paving the way for substantial advancements in quantum-based artificial intelligence applications.