HYBRID QUANTUM NEURAL NETWORK AND APPLICATION  IN WRITTEN IMAGE RECOGNITION

Trương Văn Minh; Nguyễn Minh Chiến; Phạm Sĩ Anh Đức; Nguyễn Thị Hồng; Nguyễn Hoàng Hưng Gia; Dụng Văn Lữ

doi:10.34238/tnu-jst.12645

HYBRID QUANTUM NEURAL NETWORK AND APPLICATION IN WRITTEN IMAGE RECOGNITION

About this article

Received: 21/04/25 Revised: 26/06/25 Published: 28/06/25

Authors

1. Truong Van Minh, The University of Danang - University of Science and Education
2. Nguyen Minh Chien, The University of Danang - University of Science and Education
3. Pham Si Anh Duc, The University of Danang - University of Science and Education
4. Nguyen Thi Hong, The University of Danang - University of Science and Education
5. Nguyen Hoang Hung Gia, The University of Danang - University of Science and Education
6. Dung Van Lu , The University of Danang - University of Science and Education

Abstract

Presently, a plethora of neural network architectures, including Convolutional Neural Networks, Recurrent Neural Networks, Feedforward Neural Networks… have undergone significant advancements, effectively tackling a wide array of machine learning challenges. Nevertheless, these frameworks encounter substantial obstacles when addressing tasks involving voluminous data, often resulting in suboptimal precision or necessitating protracted computational durations. Therefore, this study proposed using a hybrid quantum neural framework that amalgamates classical machine learning paradigms with quantum computation. Quantum computation augments processing velocity and elevate precision through its capacity for parallel execution and the exploitation of distinctive quantum mechanical phenomena. In this investigation, we deployed the hybrid quantum neural framework by combining two platforms the PyTorch and Qiskit on the Visual Studio Code. Experimental results show that the model achieved accuracies of 98.5% and 93.3% in recognizing handwritten digits and letters, respectively, underscoring the transformative potential of hybrid quantum neural frameworks in complex recognition applications.

Keywords

Quantum algorithm; Quantum computing; Hybrid-quantum neural network; Image recognition; PyTorch & Qiskit

Full Text:

PDF (Tiếng Việt)

References

[1] W. S. McCulloch and W. Pitts, “A logical calculus of the ideas immanent in nervous activity,” Bull. Math. Biophys., vol. 5, pp. 115–133, 1943, doi: 10.1007/BF02478259.

[2] R. Dastres and M. Soori, “Artificial neural network systems,” Int. J. Imag. Robot., vol. 21, no. 2, pp. 13–25, 2021.

[3] M. Baek et al., “Accurate prediction of protein structures and interactions using a three-track neural network,” Science, vol. 373, no. 6557, pp. 871–876, 2021, doi:10.1126/science.abj8754.

[4] J. Jumper et al., “Highly accurate protein structure prediction with AlphaFold,” Nature, vol. 596, no. 7873, pp. 583–589, 2021.

[5] D. T. Jones and J. M. Thornton, “The impact of AlphaFold2 one year on,” Nat. Methods, vol. 19, pp. 11–26, 2022.

[6] F. S. Zariquiey, R. Galvelis, E. Gallicchio, J. D. Chodera, T. E. Markland, and G. de Fabritiis, “Enhancing protein-ligand binding affinity predictions using neural network potentials,” arXiv preprint arXiv:2401.00002, 2024.

[7] J. L. Watson et al., “De novo design of protein structure and function with RFdiffusion,” Nature, vol. 620, no. 7976, pp. 1089–1100, 2023.

[8] N. Q. K. Le, V. N. Nguyen, T. T. Nguyen, T. X. Tran, and T. T. Ho, “Enhancing protein sequence classification with a fuzzy neural network: A study in anticancer peptide identification,” in Proc. 2024 Int. Conf. Fuzzy Theory Appl. (iFUZZY), Aug. 2024, pp. 1–6.

[9] N. J. Ogbuke, Y. Y. Yusuf, K. Dharma, and B. Mercangoz, “Big data supply chain analytics: Ethical, privacy and security challenges posed to business, industries and society,” Prod. Plan. Control, vol. 33, no. 2–3, pp. 123–137, 2022, doi: 10.1080/09537287.2020.1810764.

[10] M. A. Shafique, A. Munir, and I. Latif, “Quantum computing: Circuits, algorithms, and applications,” IEEE Access, vol. 12, pp. 22296–22314, 2024.

[11] H. Padmanaban, “Quantum computing and AI in the cloud,” J. Comput. Intell. Robot., vol. 4, no. 1, pp. 14–32, 2024.

[12] Y. Kwak, W. J. Yun, S. Jung, and J. Kim, “Quantum neural networks: Concepts, applications, and challenges,” in Proc. 12th Int. Conf. Ubiquitous Future Netw. (ICUFN), Aug. 2021, pp. 413–416.

[13] S. Wu, Y. Zhang, and J. Li, “Quantum data parallelism in quantum neural networks,” Phys. Rev. Res., vol. 7, no. 1, 2025, Art. no. 013177.

[14] P. N. Nguyen, “Biomarker discovery with quantum neural networks: A case-study in CTLA4-activation pathways,” BMC Bioinformatics, vol. 25, no. 1, 2024, Art. no. 149.

[15] M. Kashif, A. Marchisio, and M. Shafique, “Computational advantage in hybrid quantum neural networks: Myth or reality?,” arXiv preprint arXiv:2412.04991, 2024.

[16] M. A. Hafeez, A. Munir, and H. Ullah, “H-QNN: A hybrid quantum–classical neural network for improved binary image classification,” AI, vol. 5, no. 3, pp. 1462–1481, 2024.

[17] R. Agarwal, “Complete Guide to the Adam Optimization Algorithm,” Built In, 2023. [Online]. Available: https://builtin.com/machine-learning/adam-optimization. [Accessed March 7, 2025].

[18] Ultralytics, “MNIST Dataset,” 2023. [Online]. Available: https://docs.ultralytics.com/datasets/classify/mnist/#key-features. [Accessed Mar. 20, 2025].

[19] V. M. Truong and M. C. Nguyen, “Quantum hybrid neural network structure,” (in Vietnamese), 2024. [Online]. Available: https://drive.google.com/drive/folders/1Sh7O87q22BCZL9cdmPFjW9 fdurk2u7Ng?usp=sharing. [Accessed Jan 8, 2025].

[20] N. Ketkar and M. Masko, Deep Learning with PyTorch: A Practical Approach to Building Neural Network Models Using PyTorch. Apress, 2019.

[21] S. Patel, “A-Z handwritten alphabets in CSV format,” *Kaggle*, 2020. [Online]. Available: https://www.kaggle.com/datasets/sachinpatel21/az-handwritten-alphabets-in-csv-format. [Accessed Jan 19, 2025].

DOI: https://doi.org/10.34238/tnu-jst.12645

Refbacks

There are currently no refbacks.



Remember me