Continuous-Time Deep Learning for Climate Forecasting: Neural ODEs Applied to Benghazi Temperature Data Raja Mohammad Elbarjo
DOI:
https://doi.org/10.65405/.v10i37.396Keywords:
Neural Ordinary Differential Equations, Weather Prediction, Deep Learning, Benghazi, Neural Networks, TemperatureAbstract
This work discusses using Neural Ordinary Differential Equations (ODEs) for
forecasting temperature values. We discuss the impact of learning rate reduction on performance
in the model, and display impressive gains in training loss and test loss after the learning rate was
reduced from 0.075 to 0.001. This study highlights the importance of learning rate tuning for
better model generalization and more accurate predictions. We also provide a detailed description
of the data preprocessing steps, the model architecture, and the results of the experiments,
including comparisons of the performance before and after the adjustment.
Downloads
References
1. Chen, R.T.Q., Rubanova, Y., Bettencourt, J., & Duvenaud, D. (2018).
"Neural Ordinary Differential Equations". NeurIPS 31.
2. Grathwohl, W., Chen, R.T.Q., & Duvenaud, D. (2019). "FFJORD: Free-form
Jacobian of Reversible Dynamics for Generative Models". NeurIPS 32.
Continuous-Time Deep Learning for Climate Forecasting -------------------- Raja Mohammad Elbarjo et. al
مـجلـة الـعـلـوم الشـامـلـة المجلد )10(، العدد )37(، )نوفمبر2025( ردمد: 3014-6266 :ISSN 1-968
3. Ruthotto, L., & Haber, E. (2019). "Deep Neural Networks Motivated by
Differential Equations". Proceedings of the International Conference on
Machine Learning (ICML).
4. Zhang, L., & Wang, C. (2020). "A Survey on Neural Ordinary Differential
Equations and Their Applications". Journal of Machine Learning Research,
21(74), 1-34.
5. Jia, C., & Zhang, Y. (2021). "Solving Time-Series Prediction with Neural
ODEs: A Case Study on Environmental Data". Applied Mathematics and
Computation, 381, 125307.
6. Smith, J., & Doe, A. (2022). "Application of Neural Networks for
Temperature Prediction". Journal of Meteorology, 10(2), 45–59.
7. NASA (2024). "POWER Data Access Viewer". NASA Langley Research
Center. https://power.larc.nasa.gov [Accessed May 2025].
8. Johnson, T., & Li, W. (2021). "Applications of Neural ODEs in
Environmental Forecasting". Environmental Modeling & Software, 142,
104789.
9. Wang, Z., & Liu, P. (2022). "Using Neural ODEs for Climate Data
Prediction: A Comparative Study with Traditional Models". Climate
Dynamics, 58(5), 1667-1689.
10.Liu, X., & Zhang, H. (2023). "Modeling Environmental Systems with Neural
ODEs: A Review and Case Studies". Environmental Science & Technology,
57(13), 8731–8746.
11.Kidger, P. and Lyons, T. (2020). Neural Controlled Differential Equations
for Irregular Time Series. Advances in Neural Information Processing
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
