Neural Network by Ear: How to Capture a Particle's Energy express
Original: "Optimised neural networks for online processing of ATLAS calorimeter data on FPGAs"
A neural network built into the detector listens in on the collider's noise and measures particle energy with record accuracy.
Links in the knowledge graph 1
In a particle collider, a cacophony of signals drowns out important events, just as crowd noise hides a whisper. Engineers created a 'super-hearing' neural network that works directly in the detector, accurately measures energy, and honestly admits when it’s unsure. This brings us closer to unraveling dark matter.
🎯 The ATLAS detector is as big as an eight-story building, but the neural network processing its signals fits on a chip smaller than a matchbox.