Page 195 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

• REQ-04: It's recommended to use satellite data and radar image data to enhance
forecasting performance.
• REQ-05: It's required to use GPU for model training, which is highly recommended for
daily online inference.
• REQ-06: It's recommended to use cloud-based data transmission for deployment. Change 4.2-Climate

4 Sequence Diagram

5 References
[1] C. Liu, “Alibaba’s New AI Model Enhances Weather Forecasting Precision Amid Rising
Climate Threats,” Alizila, Nov. 06, 2024. https:// www .alizila .com/ alibaba -damo -academy
-ai -model -weather -forecasting (accessed Jun. 09, 2025).
[2] P. Niu et al., “Utilizing Strategic Pre-training to Reduce Overfitting: Baguan -- A Pre-
trained Weather Forecasting Model,” arXiv.org, 2025. https:// arxiv .org/ abs/ 2505 .13873
(accessed Jun. 09, 2025).
[3] Y. Guo, T. Zhou, W. Jiang, B. Wu, L. Sun, and R. Jin, “Maximizing the Impact of Deep
Learning on Subseasonal-to-Seasonal Climate Forecasting: The Essential Role of
Optimization,” arXiv.org, 2024. https:// arxiv .org/ abs/ 2411 .16728 (accessed Jun. 09,
2025).
[4] S. Zhang, “5 Ways Alibaba is Leading Sustainability Efforts Demonstrated at COP29,”
Alizila, Nov. 18, 2024. https:// www .alizila .com/ 5 -ways -alibaba -is -leading -sustainability
-efforts -demonstrated -at -cop29/
[5] “MindOpt_Decision Intelligence – Alibaba Cloud,” Aliyun.com, 2025. https:// opt .aliyun
.com/ weather (accessed Jun. 09, 2025).

159