Theme: The HDR ML Challenge program is hosting its second FAIR challenge for 2025, presenting three scientific benchmarks for modeling out of distribution in three critical areas: Neural Forecasting, Climate Prediction using Ecological Data, and Coastal Flooding Prediction over time. Machine learning problems are often driven by the quality of the available training datasets. Models are very effective at interpolating across their training datasets to find patterns and trends. In this challenge, we ask models to extend beyond their training by performing out of domain extrapolation to practical critical scientific process that have not yet been well studied. As with the first challenge, we will host three distinct sub-challenges on different scientific problems.
Learn More About the Three Challenges:
What’s Happening Now?
Hackathons, Datathons, and more!
- University of Maryland, Baltimore County – DATATHON (In progress)
December 1 to 15, 2025
- University of Colorado, Boulder – HACKATHON
December 12, 2025
- NSF HDR Institutes – (VIRTUAL) HACKATHON
December 18, 2025
Sign up to participate:
https://indico.cern.ch/event/1607943/registrations/participants
Upcoming Workshops!
- FAIR in ML, AI Readiness, & Reproducibility (FARR) Workshop 2026
Join us in Washington, D.C., at the AGU Conference Center for the ML Challenge Awards Ceremony
April 8-9, 2026
To learn more: https://www.farr-rcn.org/workshop26
Year 2: Sponsoring Partners
Get Involved:
Become a Sponsor!
Organize a Hackathon!
Come and Join as a Participant!
Our Year 2 Challenge is Available through January 31st, 2026
Theme: A collaborative effort among three of the five NSF HDR institutes. Each HDR institute presented an ‘Anomaly Detection‘ problem to answer questions using complex datasets in polar science, astrophysics, ecology, and evolution. This challenge was geared towards bringing awareness to the broader community about the complexity of using deep learning for this problem since the availability of well-labeled datasets or training through systematic procedures, such as masking, is often insufficient.
Learn More About The Three Challenges:
Highlights
Kick-off Hackathon Events
Approximately 10 local hackathon events to kick-off the 1st ML challenge were organized at several institutions such as University of Maryland, Baltimore County, University of Washington, and the University of California, San Diego.
AAAI 2025 Workshop and Challenge on Anomaly Detection in Scientific Domains
The workshop was co-located with The 39th Annual AAAI Conference on Artificial Intelligence. To learn more, visit https://www.nsfhdr.org/AAAI-workshop
Scholarship
Dr. Philip Harris (Director, A3D3), Massachusetts Institute of Technology (HDR Ecosystem)
FAIR in ML, AI Readiness, & Reproducibility (FARR) Workshop, AGU Conference Center in Washington D.C. (October 9 – 10, 2024)
- Invited Talk at a ML Commons virtual event held in February 2025.
K-12 Connection
iHARP’s Challenge: “Machine Learning to Predict Anomalous Flooding Events.”
iHARP Focused Outcomes: Proof-of-concept App; Virtual High School Hackathon Showcase
Institutional Organizers
The ML challenge was led by faculty, post-doctoral, and student researchers within the HDR community from the several academic institutions:
Princeton University
Year 1: Sponsoring Partners
