Scholarlink

<p>Recent advances in deep learning (DL) have significantly improved image classification techniques and systems in various domains such as healthcare, security, and autonomous systems. However, the computational complexity of classical DL models poses substantial challenges, requiring extensive resources such as large datasets and computational time for training and inference, which limits their applicability in resource-constrained environments. Quantum Machine Learning (QML) has emerged as a promising alternative leveraging the principles of quantum mechanics, such as superposition and entanglement, to facilitate efficient data processing within the Hilbert space. This work proposes the Eulerian Quantum Neural Network (EQNN), a quantum-based DL model that employs Eulerian rotational unitary matrices to achieve robust quantum feature representation using quantum gates, ensuring improved data fidelity. The proposed framework incorporates unitary operations and Pauli-Z gates to extract classical insights from quantum data, enabling seamless integration with existing classical systems. Besides computationally efficient solution for image classification, the EQNN’s performance evaluation on the MNIST, F-MNIST, and P-MNIST (Pneumonia) datasets demonstrate the model’s effectiveness in both binary and multi-class classification tasks, achieving promising results even with a limited number of trainable parameters. These findings highlight the potential of EQNN as a resource-efficient and effective approach for next-generation image classification systems. The source code of this work is available at: <a href="https://github.com/soumyadip-sahoo/EQNN">https://github.com/soumyadip-sahoo/EQNN</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Entanglement-driven learning-based angle-encoded variational quantum neural network for image classification

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Soumyadip Sahoo,
  • Diptarka Mandal,
  • Asfak Ali,
  • Dmitrii Kaplun,
  • Sergei Romanov,
  • Ram Sarkar

摘要

Recent advances in deep learning (DL) have significantly improved image classification techniques and systems in various domains such as healthcare, security, and autonomous systems. However, the computational complexity of classical DL models poses substantial challenges, requiring extensive resources such as large datasets and computational time for training and inference, which limits their applicability in resource-constrained environments. Quantum Machine Learning (QML) has emerged as a promising alternative leveraging the principles of quantum mechanics, such as superposition and entanglement, to facilitate efficient data processing within the Hilbert space. This work proposes the Eulerian Quantum Neural Network (EQNN), a quantum-based DL model that employs Eulerian rotational unitary matrices to achieve robust quantum feature representation using quantum gates, ensuring improved data fidelity. The proposed framework incorporates unitary operations and Pauli-Z gates to extract classical insights from quantum data, enabling seamless integration with existing classical systems. Besides computationally efficient solution for image classification, the EQNN’s performance evaluation on the MNIST, F-MNIST, and P-MNIST (Pneumonia) datasets demonstrate the model’s effectiveness in both binary and multi-class classification tasks, achieving promising results even with a limited number of trainable parameters. These findings highlight the potential of EQNN as a resource-efficient and effective approach for next-generation image classification systems. The source code of this work is available at: https://github.com/soumyadip-sahoo/EQNN.