Deep Learning-Based Weather Prediction: A Survey

Ren, Xiaoyong; Li, Xiaoyong; Ren, Kaijun; Song, Junqiang; Xu, Zichen; Deng, Kefeng; Wang, Xiang

doi:10.1016/j.bdr.2020.100178

Cited by 91 publications

(22 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Avg To indicate the performance of the periodicity component on both nearby-time and periodic-depth representation, we mixed the proposals of SP-CRN-1 (1,1,0), SP-CRN-2 (0,4,0), and SP-CRN-3 (0,0,2) because they performed the best in their settings from the previous experiment (Table 1). The mixed proposals were combined with all the combinations that were SP-CRN-1+2 (1,4,0), SP-CRN-1+3 (1,0,2), SP-CRN-2+3 (0,4,2), and SP-CRN-1+2+3 (1,4,2). We also experimented with the idea of a metadata component from the original Periodic-CRN [27] to indicate the month of the year with one-hot encoding.…”

Section: Methodsmentioning

confidence: 99%

“…Furthermore, it consumes a lot of computational resources to make a forecast by using numerical weather prediction (NWP) models [1], and requires climatology knowledge to build a model, unlike machine learning (ML). Currently, many applications are using ML in the climate domain, such as tropical cyclone forecasting [2,3], long-term rainfall prediction [4], and more [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Soft Periodic Convolutional Recurrent Network for Spatiotemporal Climate Forecast

et al. 2021

View full text Add to dashboard Cite

Many machine-learning applications and methods are emerging to solve problems associated with spatiotemporal climate forecasting; however, a prediction algorithm that considers only short-range sequential information may not be adequate to deal with periodic patterns such as seasonality. In this paper, we adopt a Periodic Convolutional Recurrent Network (Periodic-CRN) model to employ the periodicity component in our proposals of the periodic representation dictionary (PRD). Phase shifts and non-stationarity of periodicity are the key components in the model to support. Specifically, we propose a Soft Periodic-CRN (SP-CRN) with three proposals of utilizing periodicity components: nearby-time (PRD-1), periodic-depth (PRD-2), and periodic-depth differencing (PRD-3) representation to improve climate forecasting accuracy. We experimented on geopotential height at 300 hPa (ZH300) and sea surface temperature (SST) datasets of ERA-Interim. The results showed the superiority of PRD-1 plus or minus one month of a prior cycle to capture the phase shift. In addition, PRD-3 considered only the depth of one differencing periodic cycle (i.e., the previous year) can significantly improve the prediction accuracy of ZH300 and SST. The mixed method of PRD-1, and PRD-3 (SP-CRN-1+3) showed a competitive or slight improvement over their base models. By adding the metadata component to indicate the month with one-hot encoding to SP-CRN-1+3, the prediction result was a drastic improvement. The results showed that the proposed method could learn four years of periodicity from the data, which may relate to the El Niño–Southern Oscillation (ENSO) cycle.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Soft Periodic Convolutional Recurrent Network for Spatiotemporal Climate Forecast

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Various deep learning-based models are designed for climate and weather forecasting [1], [6], [9], [10], [33], [34], including near-surface air temperature predictions [17], [35], air quality inference [8], [36], precipitation predictions [2], [3], [37], wind speed predictions [7], [38], [39] and extreme weather predictions [4], [5]. Climate and weather consists of spatio and temporal data, leading to the development of spatio-temporal models [11], [40].…”

Section: Climate Modeling and Weather Forecastingmentioning

confidence: 99%

Climate Modeling with Neural Diffusion Equations

Hwang¹,

Choi²,

Choi³

et al. 2021

Preprint

View full text Add to dashboard Cite

Owing to the remarkable development of deep learning technology, there have been a series of efforts to build deep learning-based climate models. Whereas most of them utilize recurrent neural networks and/or graph neural networks, we design a novel climate model based on the two concepts, the neural ordinary differential equation (NODE) and the diffusion equation. Many physical processes involving a Brownian motion of particles can be described by the diffusion equation and as a result, it is widely used for modeling climate. On the other hand, neural ordinary differential equations (NODEs) are to learn a latent governing equation of ODE from data. In our presented method, we combine them into a single framework and propose a concept, called neural diffusion equation (NDE). Our NDE, equipped with the diffusion equation and one more additional neural network to model inherent uncertainty, can learn an appropriate latent governing equation that best describes a given climate dataset. In our experiments with two real-world and one synthetic datasets and eleven baselines, our method consistently outperforms existing baselines by non-trivial margins.

show abstract

“…The spatio-temporal graph data frequently happens in realworld applications, ranging from traffic to climate forecasting (Zaytar and El Amrani 2016;Shi et al 2015Shi et al , 2017Liu et al 2016;Racah et al 2016;Kurth et al 2018;Cheng et al 2018a,b;Hossain et al 2015;Ren et al 2021;Tekin et al 2021;Li et al 2018;Yu, Yin, and Zhu 2018;Wu et al 2019;Guo et al 2019;Bai et al 2019;Song et al 2020;Huang et al 2020;Bai et al 2020;Li and Zhu 2021;Chen, Segovia-Dominguez, and Gel 2021;Fang et al 2021). For instance, the traffic forecasting task launched by California Performance of Transportation (PeMS) is one of the most popular problems in the area of spatio-temporal processing (Chen et al 2001;Yu, Yin, and Zhu 2018;Guo et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Graph Neural Controlled Differential Equations for Traffic Forecasting

Choi

Hwang

et al. 2022

AAAI

View full text Add to dashboard Cite

Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this paper, we present the method of spatio-temporal graph neural controlled differential equation (STG-NCDE). Neural controlled differential equations (NCDEs) are a breakthrough concept for processing sequential data. We extend the concept and design two NCDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 20 baselines. STG-NCDE shows the best accuracy in all cases, outperforming all those 20 baselines by non-trivial margins.

show abstract

Deep Learning-Based Weather Prediction: A Survey

Cited by 91 publications

References 43 publications

Soft Periodic Convolutional Recurrent Network for Spatiotemporal Climate Forecast

Soft Periodic Convolutional Recurrent Network for Spatiotemporal Climate Forecast

Climate Modeling with Neural Diffusion Equations

Graph Neural Controlled Differential Equations for Traffic Forecasting

Contact Info

Product

Resources

About