Voir en

Are you up for the TrackML challenge?

Can machine learning assist high-energy physics in discovering and characterising new particles?

7 May, 2018

10 000 tracks grouping 100 000 points in a future LHC detector as simulated for the TrackML challenge (Image: TrackML Challenge Team/CERN)

Physicists from the ATLAS, CMS and LHCb collaborations have just launched the TrackML challenge – your chance to develop new machine-learning solutions for the next generation of particle detectors.

The Large Hadron Collider (LHC) produces hundreds of millions of collisions every second, generating tens of petabytes of data a year. Handling this flood of data is a major challenge for physicists, who have developed tools to process and filter the events online within a fraction of a second and select the most promising collision events.

Managing the amount of data will become even more challenging in the near future: a major upgrade planned to begin operation in 2026, the High-Luminosity LHC, will increase the collision rate by up to a factor of five. Innovative new software solutions will be needed to promptly reconstruct the tracks produced by these collisions with the available computing resources.

To help address this issue, a team of machine-learning experts and LHC physicists has partnered with Kaggle to probe the question: can machine learning assist high-energy physics in discovering and characterising new particles?

Specifically, this competition challenges participants to build an algorithm that quickly and efficiently reconstructs particle tracks from 3D points left in the silicon detectors. The challenge consists of two phases:

The “Accuracy Phase” is now running on Kaggle from May to July 2018. Here the focus is on the highest score, irrespective of the evaluation time. This phase is an official IEEE WCCI competition (Rio de Janeiro, July 2018).
The “Throughput Phase” will run on the Codalab platform from July to October 2018. Participants will submit their software to be evaluated by Codalab. The focus here is on the throughput (or speed) of the evaluation while also achieving a good score. This phase is an official NIPS competition (Montreal, December 2018).

Sign up for the TrackML challenge today. The three top scorers will receive cash prizes. Selected winners may be awarded a top-notch NVIDIA v100 GPU, get the chance to visit CERN or attend the 2018 Conference on Neural Information Processing Systems in Montreal (Canada).

For more information and the conditions for participation, visit the Kaggle challenge website and the official TrackML Twitter account.