TY - GEN
T1 - Climate Modeling with Neural Diffusion Equations
AU - Hwang, Jeehyun
AU - Choi, Jeongwhan
AU - Choi, Hwangyong
AU - Lee, Kookjin
AU - Lee, Dongeun
AU - Park, Noseong
N1 - Funding Information:
ACKNOWLEDGEMENT Noseong Park (noseong@yonsei.ac.kr) is the corresponding author. This work was supported by the Yonsei University Research Fund of 2021, and the Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2020-0-01361, Artificial Intelligence Graduate School Program (Yon-sei University), and No. 2021-0-00155, Context and Activity Analysis-based Solution for Safe Childcare).
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - 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.
AB - 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.
KW - climate modeling
KW - diffusion equation
KW - neural ordinary differential equation
UR - http://www.scopus.com/inward/record.url?scp=85125198008&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85125198008&partnerID=8YFLogxK
U2 - 10.1109/ICDM51629.2021.00033
DO - 10.1109/ICDM51629.2021.00033
M3 - Conference contribution
AN - SCOPUS:85125198008
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 230
EP - 239
BT - Proceedings - 21st IEEE International Conference on Data Mining, ICDM 2021
A2 - Bailey, James
A2 - Miettinen, Pauli
A2 - Koh, Yun Sing
A2 - Tao, Dacheng
A2 - Wu, Xindong
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 21st IEEE International Conference on Data Mining, ICDM 2021
Y2 - 7 December 2021 through 10 December 2021
ER -