Fair Universe HiggsML Uncertainty CERN workshop

Monday 19 May 2025, 09:00 → 13:00 Europe/Zurich

222/R-001 (CERN)

The FAIR Universe -- HiggsML Uncertainty Challenge focuses on measuring the physics properties of elementary particles with imperfect simulators.

The goal of the challenge is to bring together the physics and machine learning communities to advance our understanding and methodologies in handling systematic (epistemic) uncertainties within AI techniques.

The challenge (an official NeurIPS 2024 competition) has run 1st October 2024 till 15th March 2025, and the dataset and benchmark will remain available to evaluate further developments.

Results have just been announced !

After extensive studies on a new hold-out data set, the Vienna HEPHY team and Ibrahim Elsharkawy (U Illinois at Urbana Champaign) cannot be separated in a significant way and are declared joint first. Hashizume Yota (Kyoto University) has secured third position. 

This event is the closing workshop where the three best participants will present their solutions, as well as keynotes addressing the overall topic of uncertainty-aware training.

It is run as a satellite workshop for the 7th CERN Interexperiment Machine Learning workshop. We recommend participants to Fair Universe workshop to register here and also  for the main free IML workshop, in particular for anything concerning CERN access and accommodation

Useful links:

• Fair Universe web site and winner announcement
• Fair Universe HiggsML uncertainties competition site
• NeurIPS 2024 Fair universe workshop
• Fair universe whitepaper

 

This is a collaboration between LBNL Berkeley, University of Washington, Chalearn and University Paris-Saclay: Wahid Bhimji, Paolo Calafiura, Ragansu Chakkappai, Po-Wen Chang, Yuan-Tang Chou, Sascha Diefenbacher, Jordan Dudley, Steven Farrell, Aishik Ghosh, Isabelle Guyon, Chris Harris, Shih-Chieh Hsu, Elham E Khoda, Benjamin Nachman, Peter Nugent, David Rousseau, Benjamin Sluijter, Benjamin Thorne, Ihsan Ullah and Yulei Zhang.

09:00 → 09:30 Uncertainties in ML for HEP

Speaker: Michael Kagan (SLAC National Accelerator Laboratory (US))

Kagan_FAIR_2025.pdf

09:30 → 10:00 Systematic uncertainty-aware neural network trainings for binned-likelihood analyses at the LHC

Speaker: Roger Wolf (KIT - Karlsruhe Institute of Technology (DE))

IML2025-SANNT.pdf

10:00 → 10:30 Coffee break

10:30 → 11:00 Fair Universe HiggML uncertainty competition overview and results.

Speakers: David Rousseau (IJCLab-Orsay), Wahid Bhimji (Lawrence Berkeley National Lab. (US))

Fair Universe 2025 - IML workshop_first.pdf

11:00 → 11:15 Bronze medal talk : HZUME

Speaker: Hashizume Yota (Kyoto University)

higgsml-3rd-place.pdf

11:15 → 11:40 Gold medal tie talk - HEPHY: Unbinned inclusive cross-section measurements with machine-learned systematic uncertainties

Speaker: Robert Schoefbeck (Austrian Academy of Sciences (AT))

25-05-19 FAIR-HUC workshop GOLLUM.pdf

25-05-19 FAIR-HUC workshop GOLLUM.pptx

11:40 → 12:05 Gold medal tie talk - IBRAHIME: Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation

Speaker: Ibrahim Elsharkawy (University of Illinois Urbana-Champaign)

WorkshopPresentaion .pdf

12:05 → 12:10 Fair Universe outlook : permanent release and cosmology weak-lensing challenge

Speakers: David Rousseau (IJCLab-Orsay), Wahid Bhimji (Lawrence Berkeley National Lab. (US))

Fair Universe 2025 - IML workshop_last.pdf