Fuzzy Systems and its Applications

Fuzzy Systems and its Applications

Deep fuzzy network: A hybrid architecture for managing ambiguity in real-world data.

Document Type : Original Article

Author
Majid Darehmiraki
University of Khatamolanbia
10.22034/jfsa.2026.558048.1290
Abstract
In many real-world applications, data is ambiguous, incomplete, and noisy in nature, and using classical deep learning models based on real numbers and exact data leads to significant performance degradation. On the other hand, fuzzy logic and numbers provide powerful tools for modeling uncertainty and ambiguity in data, but their efficient integration into the structure of deep networks is still an open challenge. In this research, we propose a new architecture called deep fuzzy network, whose structure is designed based on fuzzy operations and representations. In deep fuzzy network, inputs are represented as triangular or trapezoidal fuzzy numbers, and in hidden layers, fuzzy algebra operations including fuzzy addition and multiplication are performed using membership functions. In order to control and reduce uncertainty during the learning process, a new parameter called sharpness parameter is introduced, whose role is to adjust the degree of fuzziness in the network connections. The output of the network is also interpreted as a fuzzy number that, in addition to the prediction value, also expresses the level of confidence of the model. Initial experimental results show that the proposed model is able to provide more stable and interpretable performance than traditional deep models when faced with ambiguous and qualitative data.
Keywords
Deep neural network
fuzzy numbers
fuzzy data
Ambiguity
Subjects
[1]    Behdani, Z., & Darehmiraki, M. (2024). Theil-Sen Estimators for fuzzy regression model. Iranian Journal of Fuzzy Systems, 21(3), 177-192.
[2]    Bengio, Y., Goodfellow, I., & Courville, A. (2017). Deep learning (Vol. 1, pp. 23-24). Cambridge, MA, USA: MIT press.
 
[3]    Das, R., Sen, S., & Maulik, U. (2020). A survey on fuzzy deep neural networks. ACM Computing Surveys (CSUR), 53(3), 1-25.
[4]    Farnam, M., & Darehmiraki, M. (2021). A new approach in modeling and solving the minimal spanning tree problem. Fuzzy Systems and its Applications, 4(1), 111-123.
[5]    Huang, W., Sun, M., Zhu, L., Oh, S. K., & Pedrycz, W. (2022). Deep fuzzy min–max neural network: Analysis and design. IEEE transactions on neural networks and learning systems, 35(6), 8229-8240.
[6]    Huang, Y., Chen, D., Zhao, W., & Mo, H. (2021). Deep fuzzy system algorithms based on deep learning and input sharing for regression application. International Journal of Fuzzy Systems, 23(3), 727-742.
[7]    Huang, W., Sun, M., Zhu, L., Oh, S. K., & Pedrycz, W. (2022). Deep fuzzy min–max neural network: Analysis and design. IEEE transactions on neural networks and learning systems, 35(6), 8229-8240.
[8]    Huang, Y., Chen, D., Zhao, W., & Mo, H. (2021). Deep fuzzy system algorithms based on deep learning and input sharing for regression application. International Journal of Fuzzy Systems, 23(3), 727-742.
[9]    Jang, J. S. (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665-685.
[10]    Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
[11]    Lu, H., Zhang, M., Xu, X., Li, Y., & Shen, H. T. (2020). Deep fuzzy hashing network for efficient image retrieval. IEEE transactions on fuzzy systems, 29(1), 166-176.
[12]    Price, S. R., Price, S. R., & Anderson, D. T. (2019, June). Introducing fuzzy layers for deep learning. In 2019 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) (pp. 1-6). IEEE.
[13]    Wang, G., Jia, Q. S., Qiao, J., Bi, J., & Liu, C. (2020). A sparse deep belief network with efficient fuzzy learning framework. Neural Networks, 121, 430-440.
[14]    Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). Chestx-ray8: Hospital- scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2097-2106).
 
Fuzzy Systems and its Applications
Volume 8, Issue 2 - Serial Number 17
Open Access Statement
December 2025
Pages 228-249

  • Receive Date 06 November 2025
  • Revise Date 15 February 2026
  • Accept Date 28 February 2026