We applied a variety of parametric and non-parametric machine learning models to predict the probability distribution of rainfall based on 1M training examples over a single year across several U.S. states. Our top performing model based on a squared loss objective was a cross-validated parametric k-nearest-neighbor predictor that took about six days to compute, and was competitive in a world-wide competition.
TCR-GAN: Predicting tropical cyclone passive microwave rainfall using infrared imagery via generative adversarial networks
School of Geosciences, China University of Petroleum(中国石油大学地球科学学院);College of Computer Science and Technology, China University of Petroleum(中国石油大学计算机科学与技术学院)
Analysis of Rainfall Variability and Water Extent of Selected Hydropower Reservoir Using Google Earth Engine (GEE): A Case Study from Two Tropical Countries, Sri Lanka and Vietnam
International Water Management Institute(国际水资源管理研究所)
Fusion of rain radar images and wind forecasts in a deep learning model applied to rain nowcasting
CNRS(法国国家科学研究中心);Nansen Environmental and Remote Sensing Center(南森环境与遥感中心);Laboratoire d’océanographie et du climat(海洋学与气候实验室);École nationale supérieure d'informatique pour l'industrie et l'entreprise(国立工业与企业计算机科学高等学院)
Hurdle-RMIL: Addressing Zero Inflation and Long-Tailed Imbalance in Infrared Rainfall Retrieval
Nanjing Innovation Institute for Atmospheric Sciences, Chinese Academy of Meteorological Sciences-Jiangsu Meteorological Service(南京大气科学创新研究院,中国气象科学研究院-江苏气象服务);Jiangsu Key Laboratory of Severe Storm Disaster Risk / Key Laboratory of Transportation Meteorology of CMA(江苏 severe storm disaster risk key laboratory / 中国气象局交通运输气象 key laboratory);Institute of Tibetan Plateau Meteorology, Chinese Academy of Meteorological Sciences(青藏高原气象研究所,中国气象科学研究院);Basin Key Laboratory of Sichuan Province, Institute of Tibetan Plateau Meteorology, China Meteorological Administration(四川盆地关键实验室,青藏高原气象研究所,中国气象局)