通过设计一个具有成本效益的模型和以知识为导向的模块来理解机器学习的天气预测
Minjong Cheon1,2, Jeong-Hwan Kim1, Yumi Choi1
1Climate and Environmental Research Institute, Korea Institute of Science and Technology, Seoul, South Korea.
Scientific reports
|December 14, 2025
概括
一个新的深度学习模型,KARINA,提供高效的全球天气预报与竞争力的准确性. 它利用新的模块来改进预测,同时降低计算成本,推进气象学中的机器学习.
科学领域:
- 气象学 天气学
- 人工智能的人工智能
- 计算科学 计算科学
背景情况:
- 深度学习模型越来越多地用于全球天气预报,其性能优于传统的数值模型.
- 训练这些先进的模型需要大量的计算资源,并限制了对其内部流程和精度变化的理解.
- 单个组件对模型性能和可预测性的具体贡献仍然不清楚.
研究的目的:
- 推出KARINA,一种新的数据驱动模型,用于增强天气预报,降低计算需求.
- 调查ConvNeXt骨干中的地球循环填充和SENet模块的有效性.
- 为理解机器学习天气模型机制和提高其性能提供一个框架.
主要方法:
- 开发了KARINA,这是一个基于数据的模型,集成Geocyclic Padding和SENet模块与ConvNeXt骨干.
- 进行了全面的试验,以评估KARINA的模块化组件的影响.
- 与已建立的数据驱动模型 (Pangu-Weather,GraphCast) 和数值天气预测 (ECMWF IFS) 进行了对比.
主要成果:
- 卡里纳在相对于Pangu-Weather和GraphCast的竞争性表现方面取得了较高的成绩,培训成本显著降低.
- 卡琳娜超越了ECMWF IFS数值天气预报准确度,达到10天的领先时间.
- 地气循环增强了水平向导建模,而SENet则改善了捕捉大气对流动的动态.
结论:
- 卡琳娜证明,以知识为导向的技术可以产生基于机器学习的可靠和高效的天气预测.
- 模块化设计促进了对组件贡献的理解,为未来AI天气预报方面的改进铺平了道路.
- 这项工作为深入了解模型机制提供了一条途径,并加强了高级机器学习天气预测系统的开发.
相关概念视频
01:07What is Weather?
19.6K
Overview
19.6K
01:12Precipitation Processes
4.6K
The experimental conditions in a gravimetric analysis should be optimized to maximize the particle size and purity of the obtained precipitate. Ideally, the concentration of the precipitating reagent should be low with effective stirring to maintain low relative supersaturation for the growth of large crystals. In homogeneous precipitation, the precipitant is slowly generated by a chemical reaction in the solution to avoid local reagent excesses. For example, urea decomposes gradually to...
4.6K
01:03Prediction Intervals
3.1K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
3.1K
01:30Machines: Problem Solving II
622
Machines are complex structures consisting of movable, pin-connected multi-force members that work together to transmit forces. Consider a lifting tong carrying a 100 kg load. It comprises movable sections DAF and CBG linked together with member AB.
622
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
261
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
261
01:29Mathematical Modeling: Problem Solving
218
Mathematical modeling transforms real-world scenarios into mathematical expressions, allowing for structured problem-solving and analysis. This process involves defining the situation, assigning variables to measurable quantities, selecting an appropriate model, and solving the resulting equation. Such models are invaluable in finance, providing precise methods to evaluate investments, loans, and repayment structures.A widely used example is the calculation of fixed monthly payments on a loan,...
218