Toward Foundation Models for Earth Monitoring: Proposal for a Climate Change Benchmark
Element AI;Service Now(ServiceNow);Stanford University(斯坦福大学);University of Maryland(马里兰大学);Radiant Earth Foundation;MIT(麻省理工学院);ETH Zurich(苏黎世联邦理工学院)
Alexandre Lacoste, Evan David Sherwin, Hannah Kerner, Hamed Alemohammad, Björn Lütjens, Jeremy Irvin, David Dao, Alex Chang, Mehmet Gunturkun, Alexandre Drouin,… 展开作者
Alexandre Lacoste, Evan David Sherwin, Hannah Kerner, Hamed Alemohammad, Björn Lütjens, Jeremy Irvin, David Dao, Alex Chang, Mehmet Gunturkun, Alexandre Drouin, Pau Rodriguez, David Vazquez
Recent progress in self-supervision shows that pre-training large neural networks on vast amounts of unsupervised data can lead to impressive increases in generalisation for downstream tasks. Such models, recently coined as foundation models, have been transformational to the field of natural language processing. While similar models have also been trained on large corpuses of images, they are not well suited for remote sensing data. To stimulate the development of foundation models for Earth monitoring, we propose to develop a new benchmark comprised of a variety of downstream tasks related to climate change. We believe that this can lead to substantial improvements in many existing applications and facilitate the development of new applications. This proposal is also a call for collaboration with the aim of developing a better evaluation process to mitigate potential downsides of foundation models for Earth monitoring.
Will Artificial Intelligence supersede Earth System and Climate Models?
Helmholtz Centre Potsdam, German Research Centre for Geosciences GFZ(亥姆霍兹波茨坦中心,德国地球科学研究中心GFZ);Free University of Berlin(柏林自由大学);Potsdam Institute for Climate Impact Research(波茨坦气候影响研究所);University of Exeter(埃克塞特大学);Princeton University(普林斯顿大学);NOAA/OAR Geophysical Fluid Dynamics Laboratory(美国国家海洋和大气管理局/海洋与大气研究办公室地球物理流体动力学实验室);University of Washington(华盛顿大学);Colorado State University(科罗拉多州立大学);German Climate Computing Center DKRZ(德国气候计算中心DKRZ);Helmholtz-Zentrum Geesthacht, Center for Material and Coastal Research HZG(亥姆霍兹盖斯特哈赫特中心,材料与海岸研究中心HZG)
A Self-Evolving AI Agent System for Climate Science
Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Fudan University(复旦大学);Institute of Atmospheric Physics, Chinese Academy of Sciences(中国科学院大气物理研究所);The University of Sydney(悉尼大学);Shanghai Jiao Tong University(上海交通大学);Nanjing University of Information Science and Technology(南京信息工程大学);Seoul National University(首尔大学);Ocean University of China(中国海洋大学);Laoshan Laboratory(崂山实验室);Columbia University(哥伦比亚大学);Japan Agency for Marine-Earth Science and Technology(日本海洋研究开发机构);The Chinese University of Hong Kong(香港中文大学);East China Normal University(华东师范大学);Chinese Academy of Meteorological Sciences(中国气象科学研究院);Tsinghua University(清华大学)
Fast, Scale-Adaptive, and Uncertainty-Aware Downscaling of Earth System Model Fields with Generative Machine Learning
Technical University Munich(慕尼黑工业大学);Potsdam Institute for Climate Impact Research(波茨坦气候影响研究所);Institute of Atmospheric Physics, Chinese Academy of Sciences(中国科学院大气物理研究所);University of Exeter(埃克塞特大学)
University of California, Irvine(加州大学尔湾分校);Learning the Earth with Artificial Intelligence and Physics (LEAP)(人工智能与物理学习地球中心);CMCC Foundation — Euro-Mediterranean Center on Climate Change(CMCC基金会——欧洲-地中海气候变化中心);Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR)(德国航空航天中心);University of Bremen(不来梅大学);Columbia University(哥伦比亚大学);Ecole Polytechnique(巴黎综合理工学院);University of Salento(萨伦托大学)
DeepExtremeCubes: Integrating Earth system spatio-temporal data for impact assessment of climate extremes
Leipzig University(莱比锡大学);Remote Sensing Centre for Earth System Research (RSC4Earth)(地球系统研究遥感中心);Institute for Earth System Science and Remote Sensing(地球系统科学与遥感研究所);Brockmann Consult GmbH(布罗克曼咨询有限公司);Max Planck Institute for Biogeochemistry(马克斯·普朗克生物地球化学研究所);ELLIS Unit Jena(耶拿ELLIS单元);Image Processing Laboratory (IPL)(图像处理实验室);Universitat de València(瓦伦西亚大学);Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI)(可扩展数据分析与人工智能中心)
Fusion of Heterogeneous Earth Observation Data for the Classification of Local Climate Zones
Wuhan University(武汉大学);German Aerospace Center (DLR)(德国航空航天中心);Helmholtz-Zentrum Dresden-Rossendorf (HZDR)(亥姆霍兹德累斯顿罗森多夫研究中心);Helmholtz Institute Freiberg for Resource Technology (HIF)(亥姆霍兹弗赖贝格资源技术研究所);Technical University of Munich (TUM)(慕尼黑工业大学)
ClimART: A Benchmark Dataset for Emulating Atmospheric Radiative Transfer in Weather and Climate Models
TU Darmstadt(达姆施塔特工业大学);Mila(米拉人工智能研究所);Université de Montréal(蒙特利尔大学);Environment and Climate Change Canada(加拿大环境与气候变化部);McGill University(麦吉尔大学)
地球天气和气候的数值模拟需要大量的计算。这使得人们越来越有兴趣用近似机器学习(ML)方法来取代显式计算物理过程的子程序,这些方法在推理时速度很快。在天气和气候模式中,大气辐射传输(RT)计算特别昂贵。这使得它们成为基于神经网络的仿真器的热门目标。然而,由于缺乏全面的数据集和ML基准测试的标准化最佳实践,以前的工作很难进行比较。为了填补这一空白,我们建立了一个大型数据集,ClimART,超过\emph{10 million samples from present, pre-industrial, and future climate conditions},基于加拿大地球系统模型。ClimART为ML社区带来了几个方法上的挑战,例如多个分布外测试集,底层域物理,以及准确性和推理速度之间的权衡。我们还提出了几个新的基线,表明在以前的工作中使用的数据集和网络架构的缺点。下载说明、基线和代码可从以下网址获得:https://github.com/RolnickLab/climart