Scientific Seminars

COLLOQUIUM: AI for Particle Physics: Better, Smarter, Faster

by Kevin Pedro (Fermilab)




Artificial intelligence (AI) and particle physics have a long and productive history together.
Recent advances in deep neural networks (DNNs) have led to significant improvements in the accuracy of particle identification, among other areas.  We continue to explore cutting-edge techniques such as graph neural networks, which generalize the representation of our data to exploit as much information as possible. With the coming flood of data from the High Luminosity LHC and other intensity frontier experiments, we must increase the processing speed as well as the accuracy of our algorithms.

AI satisfies both of these goals, as DNN inference can be massively accelerated using GPU or FPGA coprocessors. Connecting to coprocessors as a service minimally impacts the existing computing model, while facilitating the next generation of particle physics results.


A password is now required when joining a zoom.  People with a Fermilab  Services account can go to:

Anyone else can obtain a Zoom link by emailing Barb Kronkow at [email protected]