CCSE Retreat 2025

The 2025 CCSE Retreat was held on Friday September 19th and made possible by a grant from the Chan Wui and Yunyin Retreat Fund.
9 AM – Arrive / Breakfast
9:15 AM – Welcome Remarks
9:30 AM – Keynote: Learning global weather forecasting from data: from proof of concept to operational deployment; Remi Lam, Google DeepMind
10:30 AM – Break
10:45 AM – Session 2: Lightning Talks
12:30 PM – Lunch
1:45 PM – Sessions 3a/b, concurrently: Faculty Meeting / Career Panel
3:15 PM – Break
3:30 PM – Session 4: Small group discussions:
3:30-4:30: Modern AI Tools & CSE research
4:30-5:30: CSE education
5:30 – 6:30 PM – Reception
Talk Title: Learning global weather forecasting from data: from proof of concept to operational deployment
Speaker: Rémi Lam, Google DeepMind
Abstract: Is the future of weather forecasting written in data? This presentation will explore the recent, rapid revolution in weather prediction, where machine learning models are now learning to forecast the weather with unprecedented speed and accuracy. We will journey through the breakthroughs and the challenges of this data-driven approach. This talk will also highlight the critical pitfalls of applying machine learning in a scientific context, from the subtleties of model validation to getting adoption in the community. Finally, we’ll look to the horizon, exploring the exciting research opportunities that will define the next generation of weather prediction.
Bio: Rémi Lam is a Staff Research Scientist at Google DeepMind, where he focuses on making weather forecasting faster and more accurate. His research leverages machine learning techniques such as adversarial neural networks, graph neural networks, diffusion models, and functional generative networks to design tools for precipitation nowcasting (DGMR) and global medium-range weather prediction (GraphCast, GenCast, FGN).
GraphCast was named a runner-up for Science’s Breakthrough of the Year in 2023 and received the MacRobert Award from the Royal Academy of Engineering. For his work, Rémi was featured in the 2024 edition of Nature’s 10. Prior to joining Google DeepMind, Rémi obtained his Ph.D. in Computational Sciences and Engineering at MIT, advised by Karen Willcox.
Guest Panelists:
- Michael Brennan, Senior Researcher, Solea Energy
- Mathieu Dahan, Associate Professor, H. Milton Stewart School of Industrial and Systems Engineering, College of Engineering, Georgia Tech
- Xun Huan, Associate Professor, Mechanical Engineering, University of Michigan
- Mattan Kamon, Senior Technical Director, Lam Research
- Rémi Lam, Staff Research Scientist, Google DeepMind
- Justin Montgomery, U.S. natural gas trading expert
- Neha Sardesai, Principal Education Application Engineer, MathWorks