绿能神经架构搜索，让边缘设备高效精准预测天气
Green-NAS: A Global-Scale Multi-Objective Neural Architecture Search for Robust and Efficient Edge-Native Weather Forecasting
- 多目标搜索同时优化精度与能耗，追求低碳部署
- 模型仅15.3万参数，误差比基线高1.4%，参数量少239倍
- 适合资源有限地区，尤其数据少的城市天气预报
我们提出Green-NAS，一种面向低资源环境的多目标神经架构搜索框架，以天气预报为案例。遵循‘绿能AI’原则，该框架明确最小化计算能耗与碳足迹，优先可持续部署而非算力规模。通过联合优化多个目标，Green-NAS在保证高精度的同时，生成极轻量级模型。最佳模型Green-NAS-A仅用15.3万模型参数，达到0.0988的RMSE（较人工调优基线高1.4%），远低于全球应用模型GraphCast的参数量（减少239倍）。此外，在历史数据有限的城市，采用迁移学习可使预报精度提升约5.2%，显著优于为每城独立训练新模型的朴素方法。
原文摘要 · Abstract (English)
We introduce Green-NAS, a multi-objective NAS (neural architecture search) framework designed for low-resource environments using weather forecasting as a case study. By adhering to 'Green AI' principles, the framework explicitly minimizes computational energy costs and carbon footprints, prioritizing sustainable deployment over raw computational scale. The Green-NAS architecture search method is optimized for both model accuracy and efficiency to find lightweight models with high accuracy and very few model parameters; this is accomplished through an optimization process that simultaneously optimizes multiple objectives. Our best-performing model, Green-NAS-A, achieved an RMSE of 0.0988 (i.e., within 1.4% of our manually tuned baseline) using only 153k model parameters, which is 239 times fewer than other globally applied weather forecasting models, such as GraphCast. In addition, we also describe how the use of transfer learning will improve the weather forecasting accuracy by approximately 5.2%, in comparison to a naive approach of training a new model for each city, when there is limited historical weather data available for that city.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。