Self-learning AI separates particle jets from background noise in simulations
In short
- The AI learns on its own to separate high-energy particle jet signals from background particles.
- The method was tested on simulated relativistic heavy-ion collisions, where two ion beams meet near light speed.
- The work was led by researchers at the U.S. Department of Energy's Brookhaven National Laboratory.
Why it's good news
Heavy-ion collisions produce a huge background of particles that can hide the jets physicists want to study. The new approach learns to separate those jets without needing hand-labeled examples. The results so far come from simulations, so the method still needs testing on real collision data.
From Phys.org
A team of physicists and computer scientists led by researchers at the U.S. Department of Energy's (DOE) Brookhaven National Laboratory developed a new artificial intelligence (AI) approach that learns on its own to separate important high-energy particle jet signals from the enormous background of particles recorded in relativistic heavy-ion collisions, where two beams of heavy ions collide at near the speed of light.
Ask this story
For kids
Scientists smash tiny particles together at very high speed. The debris can be very messy to sort through. A new computer helper learns on its own how to find the important bits in the mess. Talk about it: what is one thing you would like a computer to learn to find for you?