Overview
The FAIR Universe project is a collaboration between the Lawrence Berkeley National Laboratory, Université Paris-Saclay, University of Washington and ChaLearn. The project is funded by the US Department of Energy.
The collaboration is building a large-compute-scale AI platform for the hosting of large scientific datasets, models, and machine learning competitions to drive forward discoveries in high energy physics and cosmology. The initial focus is on discovering and minimizing systematic uncertainties in High Energy Physics through a sequence of challenges that will bring together researchers across physics and machine learning.
The collaboration includes scientists from Lawrence Berkeley National Laboratory, Université Paris-Saclay, the University of Washington, and ChaLearn, a non-profit that organizes challenges to stimulate research in machine learning.
Competitions
This project has a series of competitions in the domain of High Energy Physics. It is an ongoing project and will end in 2025. Some of the organized competitions are detailed below.
1. Toy Challenge – October 2, 2023
A first toy challenge was released on the Codabench platform at the start of the project. This challenge is designed to test the platform and provide a simple example for the community to follow. A toy dataset was created to illustrate the types of analyses performed on real data from particle accelerators.
Competition website: [Link]
2. Particle Physics challenge trial – November 29, 2023
A hackathon was hosted at the Artificial Intelligence and the Uncertainty Challenge in Fundamental Physics workshop on November 29, 2023. Fair Universe released a trial version of a challenge using detailed physics simulations of accelerator data.
Competition website: [Link]
3. HiggsML Uncertainty Pilot Competition – March 12, 2024
A pilot version of HiggsML Uncertainty Competition was launched for participants. This trial competition focuses on involving participants to refine the competition by updating the task, dataset, and metric in a feedback loop.
Competition website: [Link]
4. The challenge of handling uncertainties in fundamental science @ NeurIPS 2024
A new challenge “FAIR Universe – the challenge of handling uncertainties in fundamental science” is accepted at NeurIPS 2024 Competition Track. This competition is in preparation and will be launched in August 2024.
Competition website: [Link]
Credits
- Benjamin Nachman
- Chris Harris
- Sascha Diefenbacher
- Steven Farrell
- Wahid Bhimji
- Jordan Dudley
- Elham E Khoda
- Shih-Chieh Hsu
- Yulei Zhang
- Yaun-Tang Chou
- Isabelle Guyon
- Ihsan Ullah
- David Rousseau
- Ragansu Chakkappai
- Aishik Ghosh
