APPLY HERE
LEAP, in collaboration with Columbia University’s Data Science Institute, Amazon Web Services (AWS), and NVIDIA, invites applications to participate in our January 2025 Hackathon: “Harnessing Machine Learning to Improve Subseasonal-to-Seasonal Climate Predictions.”
- Wednesday, January 15 – Thursday, January 16, 2025
- In-person at Columbia University (New York, NY)
For participants who would like to learn more about coding prior to participating in the Hackathon, please consider joining LEAP’s optional pre-Hackathon event, the Momentum Bootcamp.
HACKATHON GOAL
This Hackathon challenges teams of data science students, professionals working in the climate sector, and interested community members to build innovative demonstrations and machine-learning solutions for sub-seasonal climate modeling and prediction. Participants will use a newly released benchmarking dataset, ChaosBench (please read here and here for a description), to:
- illustrate the skill and limitations of current predictive tools,
- explore the value of such predictions for downstream applications, or
- improve current models by integrating machine learning, physics, and other domain knowledge.
Teams can supplement this dataset with additional data sources to enrich their project. Projects will be evaluated on real-world relevance, innovative integration of machine learning and domain science, clear presentation, and the effective use of external data.
HACKATHON ELIGIBILITY
This Hackathon is open to participants at all skill levels and from a range of disciplines and data needs in support of LEAP’s commitment to broadening participation in climate data science and making LEAP data and code broadly accessible. We aim to support travel to the Hackathon for a limited number of participants from the Global South and areas most gravely impacted by the climate crisis.
