AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model

Xingjian Shi, Zhihan Gao, Leonard Lausen, Hao Wang, Dit-Yan Yeung, Wai-kin Wong, Wang-chun WOO

NeurIPS 2017 · Architectures and efficiency › CNN design · approach: Compact CNN architectures · 612 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace fixed convolutional recurrent connections with learned, location-varying trajectoriesConvGRU → TrajGRU · HKO-7 offlineB-MSE: 5951 -> 5816Improved
replace fixed convolutional recurrent connections with learned, location-varying trajectoriesConvGRU → TrajGRU · HKO-7 online, r ≥ 30CSI: 0.1837 -> 0.1933Improved
train ConvGRU with balanced losses rather than original MSE and MAE (ablation)ConvGRU · HKO-7 offlineB-MSE: 9087 -> 5951Improved
train ConvGRU with balanced losses rather than original MSE and MAE (ablation)ConvGRU · HKO-7 offline, r ≥ 30CSI: 0.0712 -> 0.1776Improved
enable online fine-tuningTrajGRU · HKO-7, r ≥ 30CSI: 0.1856 -> 0.1933Improved
increase learned recurrent links from 5 to 13 (ablation)TrajGRU · MovingMNIST++ test setTest MSE × 10⁻²: 1.351 -> 1.170Improved
use DFN local-filter prediction instead of TrajGRUMovingMNIST++ video prediction · MovingMNIST++ test setTest MSE × 10⁻²: 1.138 -> 1.461Hurt
use 3D rather than 2D CNNCNN video prediction · HKO-7 offlineB-MSE: 7332 -> 7202Improved
use balanced losses instead of conventional unweighted lossesprecipitation nowcasting loss evaluation · HKO-7 benchmarkcorrelation with CSI and HSSImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy