用自监督学习提升降水短时预报精度，无需大量标注数据。
Self-supervised Spatial-Temporal Learner for Precipitation Nowcasting
- 通过掩码图像建模预训练编码器-解码器结构，学习时空特征。
- 在NL-50数据集上优于SmaAt-UNet，6小时内预测更准确。
- 适合需要低标注成本的气象预报场景，尤其适用于数据稀缺地区。
短时天气预报对及时决策至关重要，其中降水预报需在6小时内完成局部区域预测。该任务可视为时空序列预测问题，深度学习方法表现优异。然而，尽管自监督学习进展显著，多数成功模型仍依赖全监督。自监督学习能有效减少对标注数据的依赖。本文提出一种新模型SpaT-SparK，结合自监督与时空学习：使用基于CNN的编码器-解码器结构，在掩码图像建模（MIM）任务上预训练，再通过翻译网络捕捉历史与未来降水图之间的时序关系。在NL-50数据集上的实验表明，SpaT-SparK优于现有监督基线模型（如SmaAt-UNet），提供更精准的短时预报结果。
原文摘要 · Abstract (English)
Nowcasting, the short-term prediction of weather, is essential for making timely and weather-dependent decisions. Specifically, precipitation nowcasting aims to predict precipitation at a local level within a 6-hour time frame. This task can be framed as a spatial-temporal sequence forecasting problem, where deep learning methods have been particularly effective. However, despite advancements in self-supervised learning, most successful methods for nowcasting remain fully supervised. Self-supervised learning is advantageous for pretraining models to learn representations without requiring extensive labeled data. In this work, we leverage the benefits of self-supervised learning and integrate it with spatial-temporal learning to develop a novel model, SpaT-SparK. SpaT-SparK comprises a CNN-based encoder-decoder structure pretrained with a masked image modeling (MIM) task and a translation network that captures temporal relationships among past and future precipitation maps in downstream tasks. We conducted experiments on the NL-50 dataset to evaluate the performance of SpaT-SparK. The results demonstrate that SpaT-SparK outperforms existing baseline supervised models, such as SmaAt-UNet, providing more accurate nowcasting predictions.
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。