mlsanityData to 14 Oct 2026

MS-nowcasting: Operational Precipitation Nowcasting with Convolutional LSTMs at Microsoft Weather

arXiv 2111.09954 ↗First posted 18 Nov 2021cs.LG physics.ao-phcited by 12

Not on the map. Why?

Abstract (arXiv)

We present the encoder-forecaster convolutional long short-term memory (LSTM) deep-learning model that powers Microsoft Weather's operational precipitation nowcasting product. This model takes as input a sequence of weather radar mosaics and deterministically predicts future radar reflectivity at lead times up to 6 hours. By stacking a large input receptive field along the feature dimension and conditioning the model's forecaster with predictions from the physics-based High Resolution Rapid Refresh (HRRR) model, we are able to outperform optical flow and HRRR baselines by 20-25% on multiple metrics averaged over all lead times.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 9 of its 18 references that are in the map. First posted on arXiv 18 Nov 2021.

Builds on

9 of 18 references.

References · 9

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,436
  3. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,209
  4. 2018Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNscited by 588
  5. 2019Axial Attention in Multidimensional Transformerscited by 358
  6. 2021Skillful Precipitation Nowcasting using Deep Generative Models of Radarcited by 311
  7. 2017Deep Learning for Precipitation Nowcasting: A Benchmark and A New Modelcited by 182
  8. 2020MetNet: A Neural Weather Model for Precipitation Forecastingcited by 131
  9. 2019Precipitation Nowcasting with Satellite Imagerycited by 18

Built on by

Cited by year

Citing papers per year: 2022: 5; 2023: 4; 2024: 2; 2025: 1.

12 papers.

  1. 2022Forecasting Global Weather with Graph Neural Networkscited by 168
  2. 2022A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecastscited by 66
  3. 2023DeepSpeed4Science Initiative: Enabling Large-Scale Scientific Discovery through Sophisticated AI System Technologiescited by 15
  4. 2023TempEE: Temporal-Spatial Parallel Transformer for Radar Echo Extrapolation Beyond Auto-Regressioncited by 12
  5. 2022Domain Generalization Strategy to Train Classifiers Robust to Spatial-Temporal Shiftcited by 7
  6. 2024Extreme Precipitation Nowcasting using Transformer-based Generative Modelscited by 6
  7. 2022Generative Modeling of High-resolution Global Precipitation Forecastscited by 3
  8. 2022RainUNet for Super-Resolution Rain Movie Prediction under Spatio-temporal Shiftscited by 3
  9. 2023Inductive biases in deep learning models for weather predictioncited by 3
  10. 2023Learning Robust Precipitation Forecaster by Temporal Frame Interpolationcited by 1
  11. 2024An ensemble of data-driven weather prediction models for operational sub-seasonal forecastingcited by 1
  12. 2025BlockGPT: Spatio-Temporal Modelling of Rainfall via Frame-Level Autoregressioncited by 1