Precipitation Nowcasting with Star-Bridge Networks

TL;DR

This paper introduces StarBriNet, a novel RNN-based network with multiple sub-networks, star-shaped information bridges, and a multi-sigmoid loss function, significantly improving short-term rainfall prediction accuracy.

Contribution

The paper presents a new RNN architecture with specialized sub-networks and innovative information flow mechanisms for more accurate precipitation nowcasting.

Findings

Outperforms existing algorithms on radar echo dataset

Achieves higher prediction accuracy for various rainfall intensities

Demonstrates the effectiveness of star-shaped information bridges

Abstract

Precipitation nowcasting, which aims to precisely predict the short-term rainfall intensity of a local region, is gaining increasing attention in the artificial intelligence community. Existing deep learning-based algorithms use a single network to process various rainfall intensities together, compromising the predictive accuracy. Therefore, this paper proposes a novel recurrent neural network (RNN) based star-bridge network (StarBriNet) for precipitation nowcasting. The novelty of this work lies in the following three aspects. First, the proposed network comprises multiple sub-networks to deal with different rainfall intensities and duration separately, which can significantly improve the model performance. Second, we propose a star-shaped information bridge to enhance the information flow across RNN layers. Third, we introduce a multi-sigmoid loss function to take the precipitation nowcasting criterion into account. Experimental results demonstrate superior performance for precipitation nowcasting over existing algorithms, including the state-of-the-art one, on a natural radar echo dataset.