Meta-learning Driven Automatic Hyperparameter Optimization for Neural Networks in Computer Vision

Main Article Content

Xin Zhang
School of Landscape Architecture, Beijing Forestry University, Beijing 100083, China

DOI: https://doi.org/10.70267/cai.25v2n3.1019

Keywords

meta-learning, computer vision, hyperparameter optimization, MAML, reptile

Abstract

With the growing complexity of computer vision tasks and the explosive expansion of data scales, the optimization of neural network hyperparameters has become increasingly critical to model performance. Efficient hyperparameter optimization enables breakthroughs in tackling complex tasks and adapting to intricate scenarios. Traditional optimization methods rely heavily on manual experience and suffer from low efficiency. In contrast, meta-learning empowers hyperparameter optimization through its core logic of “learning to learn”: it extracts generalizable experience to build meta-cognition, thereby enhancing computational efficiency while achieving strong generalization capabilities. This paper first systematically introduces basic concepts such as neural networks, then elaborates on classic traditional optimization methods and meta-learning-based approaches, and then presents the experimental results of selected algorithms. Finally, it identifies unresolved issues and provides an outlook on future development trends.

Abstract 0 | PDF Downloads 0

References

  • Bello, I., Zoph, B., Vasudevan, V., & Le, Q. V. (2017). Neural optimizer search with reinforcement learning [Paper presentation]. Proceedings of the 34 th International Conference on Machine Learning, Sydney, Australia.
  • Chen, S. P., Wu, J., & Chen, X. Y. (2020). Hyperparameter optimization methods based on reinforcement learning. Mini-micro Computer Systems, 41(4), 679-684.
  • Chen, Y., Hoffman, M. W., Colmenarejo, S. G., Denil, M., Lillicrap, T. P., Botvinick, M., & Freitas, N. (2017). Learning to learn without gradient descent by gradient descent [Paper presentation]. Proceedings of the 34 th International Conference on Machine Learning, Sydney, Australia.
  • Deng, J. (2023). Research on open-world knowledge graph completion based on meta-learning [Master's thesis, Jilin University]. CNKI. https://doi.org/10.27162/d.cnki.gjlin.2023.005867.
  • Deng, S. (2019). Hyperparameter optimization method of CNN based on improved Bayesian optimization algorithm. Application Research of Computers, 36(7), 1984-1987. https://doi.org/10.19734/j.issn.1001-3695.2018.01.0021
  • Deng, T. Y., & Zhang, G. P. (2024). Research on improving small sample model generalization performance based on meta-learning and data augmentation. Modern Information Technology, 8(8), 93-96. https://doi.org/10.19850/j.cnki.2096-4706.2024.08.021
  • Dou, J. (2022). Application research on computer virus classification based on adaptive N-gram algorithm and MAML model [Master's thesis, Hainan Normal University]. CNKI. https://doi.org/10.27719/d.cnki.ghnsf.2022.000288
  • Eliasmith, C., & Anderson, C. H. (2003). Neural engineering: Computation, representation, and dynamics in neurobiological systems. MIT Press.
  • Fan, X. (2021). Research on pedestrian re-identification based on multi-task learning [Master's thesis, Nanjing University of Posts and Telecommunications]. CNKI. https://doi.org/10.27251/d.cnki.gnjdc.2021.000407
  • Finn, C., Abbeel, P., & Levine, S. (2017). Model-agnostic meta-learning for fast adaptation of deep networks [Paper presentation]. Proceedings of the 34 th International Conference on Machine Learning, Sydney, Australia.
  • Gao, Z. (2023). Research on automatic classification method of skin lesions based on deep meta-learning [Master's thesis, Shandong Normal University]. https://doi.org/10.27280/d.cnki.gsdsu.2023.000695
  • Hu, H., & Wang, J. (2025). Countermeasures for the application of computer vision image processing technology. Digital Design, 1, 42-44.
  • Huang, G., Zhu, Q., & Siew, C.-K. (2004, 25-29 July 2004). Extreme learning machine: a new learning scheme of feedforward neural networks [Paper presentation]. 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541), Budapest, Hungary.
  • Ji, C., Gao, Z., & Qin, J. (2022). A review of image classification algorithms based on convolutional neural networks. Computer Applications, 42(4), 1044-1049. https://doi.org/10.11772/j.issn.1001-9081.2021071273
  • Lee, J., Seo, H., Choi, Y. J., Lee, C., Kim, S., Lee, Y. S., Lee, S., & Kim, E. (2023). An endodontic forecasting model based on the analysis of preoperative dental radiographs: A pilot study on an endodontic predictive deep neural network. Journal of Endodontics, 49(6), 710-719. https://doi.org/10.1016/J.JOEN.2023.03.015
  • Li, F. C., Liu, Y., Wu, P. X., 董方, 蔡奇, & 王哲. (2021). A review of meta-learning research. Journal of Computer Research and Development, 44(2), 422-446.
  • Li, H. X., Song, D. L., Kong, J. N., & et al. (2024). Evaluation of hyperparameter optimization techniques for traditional machine learning models. Computer Science, 52(8), 242-255.
  • Li, J. (2025). Research on cardiac medical image segmentation method based on CNN-transformer hybrid network [Master's thesis, Qilu University of Technology]. CNKI. https://doi.org/10.27278/d.cnki.gsdqc.2025.000715
  • Li, Y. R., Zhang, Y. L., & Wang, J. C. (2022). A review of Bayesian optimization methods for hyperparameter estimation. Computer Science, 49(S1), 86-92.
  • Li, Z., Zhou, F., Chen, F., & Li, H. (2017). Meta-SGD: Learning to learn quickly for few-shot learning. arXiv preprint, arXiv:1707.09835. https://doi.org/10.48550/arXiv.1707.09835
  • Liu, B. (2025). Development and application of artificial intelligence computer vision technology in the era of big data. Information and Computer, 31(5), 34-36.
  • Liu, X. (2022). Research on automatic hyperparameter optimization based on reinforcement learning and meta-learning [Master's thesis, University of Electronic Science and Technology of China]. CNKI. https://doi.org/10.27005/d.cnki.gdzku.2022.001575
  • Liu, Y. (2022). Research on automatic hyperparameter optimization methods based on cognitive meta-learning [Master's thesis, Nanjing University of Aeronautics and Astronautics]. CNKI. https://doi.org/10.27239/d.cnki.gnhhu.2022.001508
  • Liu, Z. (2019). On computer vision technology. Digital User, 25(8), 159.
  • Long, M., & Wang, J. (2015). Learning multiple tasks with deep relationship networks. arXiv preprint, arXiv:1506.02117. https://doi.org/10.48550/arXiv.1506.02117
  • Miao, J. (2025). Research on convolutional neural network optimization and application based on transformer and collaborative attention mechanism [Master's thesis, University of Electronic Science and Technology of China]. CNKI. https://doi.org/10.27005/d.cnki.gdzku.2025.002766
  • Nigat, T. D., Sitote, T. M., & Gedefaw, B. M. (2023). Fungal skin disease classification using the convolutional neural network. Journal of Healthcare Engineering, 2023, Article 6370416. https://doi.org/10.1155/2023/6370416
  • Pan, C., & Zhang, C. (2005). Brief introduction to computer vision. Automation Exposition, 22(5), 92-93,99.
  • Pang, H. Y. (2023). Research on personalized recommendation algorithm based on meta-learning [Master's thesis, Jilin University]. CNKI. https://doi.org/10.27162/d.cnki.gjlin.2023.002492
  • Pang, Y., Xu, H., Zhang, Y., Zhu, H. L., & Peng, X. (2023). Modulation recognition algorithm based on transfer meta-learning. Acta Armamentarii, 44(10), 2954-2963. https://doi.org/10.12382/bgxb.2022.0583
  • Peng, J. J. (2025). Hyperparameter optimization research on Bayesian optimization algorithm based on GP. China New Technologies and Products, (11), 38-40. https://doi.org/10.13612/j.cnki.cntp.2025.11.047
  • Playout, C., Duval, R., & Cheriet, F. (2019). A Novel Weakly Supervised Multitask Architecture for Retinal Lesions Segmentation on Fundus Images. IEEE Transactions on Medical Imaging, 38(10), 2434-2444. https://doi.org/10.1109/TMI.2019.2906319
  • Ren, H., & Wang, X. (2021). Review of attention mechanisms. Computer Applications, 41(S1), 1-6.
  • Ren, Z., & Lee, Y. J. (2018, 18-23 June 2018). Cross-domain self-supervised multi-task feature learning using synthetic imagery [Paper presentation]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.
  • Shang, Q., Zhou, L., & Feng, L. (2019). Multi-task optimization algorithm based on denoising autoencoder. Journal of Dalian University of Technology, 59(4), 417-426. https://doi.org/10.7511/dllgxb201904013
  • Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., Van Den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., Dieleman, S., Grewe, D., Nham, J., Kalchbrenner, N., Sutskever, I., Lillicrap, T., Leach, M., Kavukcuoglu, K., Graepel, T., & Hassabis, D. (2016). Mastering the game of Go with deep neural networks and tree search. Nature, 529(7587), 484-489. https://doi.org/10.1038/NATURE16961
  • Snoek, J., Larochelle, H., & Adams, R. P. (2012). Practical bayesian optimization of machine learning algorithms [Paper presentation]. Advances in neural information processing systems, Lake Tahoe, NE, USA.
  • Tong, W. G., Li, M. X., & Zhang, Y. K. (2018). Research on deep learning optimization algorithms. Computer Science, 45(S2), 155-159.
  • Vilalta, R., & Drissi, Y. (2002). A perspective view and survey of meta-learning. Artificial Intelligence Review, 18(2), 77-95. https://doi.org/10.1023/A:1019956318069
  • Xia, Y., & Cui, W. (2022). Personalized federated learning algorithm based on Reptile. Computer Systems Applications, 31(12), 294-300. https://doi.org/10.15888/j.cnki.csa.008875
  • Xie, C. (2024). Research on few-shot learning image classification algorithms [Master's thesis, Guizhou University of Finance and Economics]. CNKI. https://doi.org/10.27731/d.cnki.ggzcj.2024.000162
  • Ye, J. (2022). Research on image pixel-level multi-vision task learning [Master's thesis, Jiangxi University of Science and Technology]. CNKI. https://doi.org/10.27176/d.cnki.gnfyc.2022.000531.
  • Yi, D. (2021). Research on YOLO-based target detection optimization algorithm [Master's thesis, Nanjing University of Posts and Telecommunications]. CNKI. https://doi.org/10.27251/d.cnki.gnjdc.2021.000149
  • Zhang, S., Gong, Y., & Wang, J. (2019). Development of deep convolutional neural networks and their applications in computer vision. Journal of Computer Science, 42(3), 453-482. https://doi.org/10.11897/SP.J.1016.2019.00453
  • Zhang, Y., & Yang, Q. (2022). A survey on multi-task learning. IEEE Transactions on Knowledge and Data Engineering, 34(12), 5586-5609. https://doi.org/10.1109/TKDE.2021.3070203

Article Sidebar

PDF
Published
Sep 12, 2025
Issue
Vol. 2 No. 3 (2025)
Section
Research Articles
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Zhang, X. . (2025). Meta-learning Driven Automatic Hyperparameter Optimization for Neural Networks in Computer Vision. Computers and Artificial Intelligence, 2(3), 10-19. https://doi.org/10.70267/cai.25v2n3.1019

Similar Articles

1-10 of 13

You may also start an advanced similarity search for this article.