Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Earthformer: Exploring Space-Time Transformers for Earth System Forecasting

About

Conventionally, Earth system (e.g., weather and climate) forecasting relies on numerical simulation with complex physical models and are hence both expensive in computation and demanding on domain expertise. With the explosive growth of the spatiotemporal Earth observation data in the past decade, data-driven models that apply Deep Learning (DL) are demonstrating impressive potential for various Earth system forecasting tasks. The Transformer as an emerging DL architecture, despite its broad success in other domains, has limited adoption in this area. In this paper, we propose Earthformer, a space-time Transformer for Earth system forecasting. Earthformer is based on a generic, flexible and efficient space-time attention block, named Cuboid Attention. The idea is to decompose the data into cuboids and apply cuboid-level self-attention in parallel. These cuboids are further connected with a collection of global vectors. We conduct experiments on the MovingMNIST dataset and a newly proposed chaotic N-body MNIST dataset to verify the effectiveness of cuboid attention and figure out the best design of Earthformer. Experiments on two real-world benchmarks about precipitation nowcasting and El Nino/Southern Oscillation (ENSO) forecasting show Earthformer achieves state-of-the-art performance. Code is available: https://github.com/amazon-science/earth-forecasting-transformer .

Zhihan Gao, Xingjian Shi, Hao Wang, Yi Zhu, Yuyang Wang, Mu Li, Dit-Yan Yeung• 2022

Related benchmarks

TaskDatasetResultRank
Video PredictionMoving MNIST
SSIM0.8961
52
Precipitation forecastingSEVIR (test)
CSI (16)76.82
34
Radar NowcastingSEVIR
SSIM72.9
31
Precipitation nowcastingMeteoNet
SSIM0.7806
29
Weather forecastingSEVIR
CSI (M)44.73
20
Precipitation nowcastingSEVIR
CSI (M)28
20
Climate ForecastingICAR-ENSO (test)
C-Niño3.4-M0.7329
17
Precipitation nowcastingARSO (test)
CSI-M40.3
14
Vegetation ForecastingGreenEarthNet 25-day lead time (OOD-t)
RMSE0.09
14
Vegetation ForecastingGreenEarthNet OOD-t 100-day lead time
R20.52
14
Showing 10 of 26 rows

Other info

Follow for update