Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
This addresses a crucial weather forecasting problem for meteorology and public safety, offering a novel machine learning approach with demonstrated improvements.
The paper tackles precipitation nowcasting by formulating it as a spatiotemporal sequence forecasting problem and proposes the ConvLSTM network, which outperforms FC-LSTM and the state-of-the-art ROVER algorithm in experiments.
The goal of precipitation nowcasting is to predict the future rainfall intensity in a local region over a relatively short period of time. Very few previous studies have examined this crucial and challenging weather forecasting problem from the machine learning perspective. In this paper, we formulate precipitation nowcasting as a spatiotemporal sequence forecasting problem in which both the input and the prediction target are spatiotemporal sequences. By extending the fully connected LSTM (FC-LSTM) to have convolutional structures in both the input-to-state and state-to-state transitions, we propose the convolutional LSTM (ConvLSTM) and use it to build an end-to-end trainable model for the precipitation nowcasting problem. Experiments show that our ConvLSTM network captures spatiotemporal correlations better and consistently outperforms FC-LSTM and the state-of-the-art operational ROVER algorithm for precipitation nowcasting.