The European Center for Nuclear Research (European Nuclear Research Center) has launched an online contest calling on stakeholders to recognize the Higgs bosoms among a large number of dataCERN).
The competition, which has been published in a known for this purpose site works as follows: participants are asked to process data from 800.000 particle collisions produced by a LHC Large Hadron Accelerator simulator.
The data can be divided into two categories: either as those in which the Higgs particle appears to be involved, or those that Higgs does not participate in. Each of the collisions is accompanied by 30 variables, such as the energy and direction of the particles that result from them.
Of the 800.000 collisions, 250.000 have already been labeled as 'Higgs' or 'not Higgs' and based on this information, the contestants are asked to 'train' a program computer so as to distinguish the rest between the two categories.
Participants then upload their results to the website which evaluates them by compiling a ranking list. This machine learning competition is accompanied by cash prizes that in total exceed 10.000 euros and to date more than 1.100 teams have submitted participation.
The team that will finish first will also be invited to the CERN research center to discuss with the scientists about its solution and to check if it could be adapted to real research needs.
Machine learning plays an ever-increasing role in modern research, and particular methods proved very useful for the discovery of the Higgs boson in 2012. Its key feature is the “training” of aletterthe computer based on a set of data so that it can then figure out for itself what to do.
"People love these kinds of problems, they capture their imaginations," said Joyce Noah-Vanuke, an analyst at contest organizer Web.σελίδαs. "However, we strive to be the online home of data analytics."
The competition has so far attracted a diverse mix of people of different skill sets as among the first-timers so far are people with no professional connection to the physics. It will accept entries until the end of September.
