Climate-AI
Climate-AI
Embedded Files
As an advocate of climate actions for the climate change, I have a strong passion in applying various AI techniques for climate domains, specifically for the climate intervention, and climate risk mitigation. I have been working on various climate AI projects so far in LLNL and PARC, and have been served as a programming or organizing committee in various climate-AI workshops held in big ML venues, such as ICML, ICLR, Neurips and ICDM.
Applying AI techniques to help to resolve climate problem is one of the big missions of my life. As most climate problem can be tackled by directly applying machine perception and vision-based representation learning techniques, there are lots of rooms to be improved in many climate problems by taking advantages of the state-or-the-art ML methods. Any discussions and collaboration opportunities are welcomed!
As an advocate of climate actions for the climate change, I have a strong passion in applying various AI techniques for climate domains, specifically for the climate intervention, and climate risk mitigation. I have been working on various climate AI projects so far in LLNL and PARC, and have been served as a programming or organizing committee in various climate-AI workshops held in big ML venues, such as ICML, ICLR, Neurips and ICDM.
Applying AI techniques to help to resolve climate problem is one of the big missions of my life. As most climate problem can be tackled by directly applying machine perception and vision-based representation learning techniques, there are lots of rooms to be improved in many climate problems by taking advantages of the state-or-the-art ML methods. Any discussions and collaboration opportunities are welcomed!
Tracking, Predicting Extreme Climate Events using video prediction models
1. Developing hurricane tracking and prediction models built upon ConvLSTM.
2. Resolved the limited labeled data problem in hurricane tracking by adopting data augmentation techniques using conditional-GAN
3. Apply Neural-ODE to interpolate and extrapolate Climate-video containing hurricane which are irregularly sampled in time-domain.
Tracking, Predicting Extreme Climate Events using video prediction models
1. Developing hurricane tracking and prediction models built upon ConvLSTM.
2. Resolved the limited labeled data problem in hurricane tracking by adopting data augmentation techniques using conditional-GAN
3. Apply Neural-ODE to interpolate and extrapolate Climate-video containing hurricane which are irregularly sampled in time-domain.
Applying Representation learning techniques for various climate problems
Applying Representation learning techniques for various climate problems
Super-resolution techniques for improve resolution of numerical weather prediction.
Temporal prediction model for predicting concentrations of future air-pollutants.
Selected Publications (* Equal contribution)
Kangyeol Kim,* Sunghyun Park,* Junsoo Lee, Joonseok Lee, Sookyung Kim, Jaegul Choo, Edward Choi, "Continuous-Time Video Generation via Learning Motion Dynamics with Neural ODE". the 32nd British Machine Vision Conference (BMVC), 2021.
Sunghyun Park, Kangyeol Kim, Junsoo Lee, Jaegul Choo, Joonseok Lee, Sookyung Kim, Edward Choi. "Vid-ODE: Continuous-Time Video Generation with Neural Ordinary Differential Equation". Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI), 2021.
Ahreum Lee, Sujong Jeong, Jaewon Joo, Chan-Ryul Park, Jhoon Kim, Sookyung Kim. "Potential role of urban forest in removing PM2.5: A case study in Seoul by deep learning with satellite data". Urban Climate, 2021.
Sookyung Kim*, Sunghyun Park*, Sunghyo-Chung*, Yun-sung Lee, Joonseok Lee, Hyojin Kim, Mr Parbhat, Jaegul Choo. "Learning to Focus and Track Extreme Climate Events". British Machine Vision Conference (BMVC) (Spotlight Talk), 2019. (*equal contribution)
Sookyung Kim, Hyojin Kim, Sangwoong Yoon, Joonseok Lee, Samira Kahou, Karthik Kashinath, Mr Prabhat. "Deep-Hurricane-Tracker: Tracking and Predicting Extreme Climate Events using ConvLSTM". IEEE Winter Conference on Applications of Computer Vision (WACV), 2019.
Others (Workshop, Talks, Seminars) (* Equal contribution)
Md Rifat Arefin, Vincent Michalski, Pierre-Luc St-Charles, Alfredo Kalaitzis, Sookyung Kim, Samira E Kahou, Yoshua Bengio. "Multi-Image Super-Resolution for Remote Sensing Using Deep Recurrent Networks". Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2020.
Sunghyun Park*, Kangyeol Kim*, Sookyung Kim*, Joonseok Lee, Junsoo Lee, Jiwoo Lee, Jaegul Choo. "Hurricane Nowcasting with Irregular Time-step using Neural-ODE and Video Prediction". Tackling Climate Change with Machine Learning on ICLR, 2020.
Sookyung Kim*, Sunghyun Park*, Sunghyo-Chung*, Yun-sung Lee, Joonseok Lee, Hyojin Kim, Jaegul Choo, Mr Parbhat. "Focus and Track: Pixel-wise Spatio-temporal Hurricane Tracking". Climate Change: How AI can help on International Conference on Machine Learning (ICML), 2019.
Sookyung Kim, Sangwoong Yoon, Hyojin Kim, Samira Kahou, Karthik Kashinath, Mr Prabhat. "Tracking Extreme Climate Events in Spatio-temporal Climate Data". European Conference on Computer Vision (ECCV) WiCV Workshop, 2018.
Sookyung Kim, Sangwoong Yoon, Hyojin Kim, Samira Kahou, Karthik Kashinath, Mr Prabhat. "Tracking extreme climate events using Neural Networks". Climate Informatics (with spotlight oral presentation), 2018.
Sookyung Kim, Jungmin. M. Lee, Jiwoo Lee, and Jihoon Seo. "Deep-dust: Predicting concentrations of fine dust in Seoul using LSTM". Climate Informatics, 2018.
Sookyung Kim, Mayur Mudigonda, Ankur Mahesh, Samira Kahou, Karthik Kashinath, Dean Williams, Vincent Michalski, Travis O'Brien, Mr Prabhat. "Segmenting and Tracking Extreme Climate Events using Neural Networks". Deep Learning for Physical Science Workshop on NIPS, 2017.
Sookyung Kim, Sasha Ames, Jiwoo Lee, Chengzhu Zhang, Aaron C. Wilson, Dean Williams. "Framework for Detection and Localization of Extreme Climate Event with Pixel Recursive Super Resolution". 7th IEEE ICDM Workshop on Data Mining on Earth System Science, 2017.
Sookyung Kim, Sasha Ames, Jiwoo Lee, Chengzhu Zhang, Aaron C.Wilson, Dean Williams. "Massive Scale Deep Learning For Detecting Extreme Climate Events". Climate Informatics, 2017.
Sookyung Kim. Machine Learning for Earth System Grid Federation (ESGF). SGF Proposal Review Meeting, 2017.
Sookyung Kim. Massive Scale Deep Learning for Predicting Extreme Climate Events. Uncertainty Quantification and Data-Driven Modeling, 2017.
AI-assisted Climate Tipping-point Modeling (AiBedo)
AiBeDO is part of DARPA ACTM project. As a technical leader of the project, developed the hybrid AI model framework using Spherical Graph Neural-Netwroks and Spherical-ConvLSTM to resolve the weaknesses of Earth System Models by generating rapid and robust multi-decadal climate projections.
AI-assisted Climate Tipping-point Modeling (AiBedo)
AiBeDO is part of DARPA ACTM project. As a technical leader of the project, developed the hybrid AI model framework using Spherical Graph Neural-Netwroks and Spherical-ConvLSTM to resolve the weaknesses of Earth System Models by generating rapid and robust multi-decadal climate projections.
2. Project page:
2. Project page:
Selected Publications
Sookyung Kim et al. "AiBEDO: A hybrid AI model to capture the effects of cloud properties on global circulation and regional climate patterns. AGU (American Geophysical Union), 2022. [Oral Presentation]
Salva R Cachay, Peetak Mitra, Sookyung Kim, Subhashis Hazarika, Kalai Ramea. "ClimFormer - a Spherical Transformer model for long-term climate projections". Submitted to the Machine Learning and the Physical Sciences Workshop at NeurIPS, 2023.
Page updated
Google Sites
Report abuse