Dortmund 2021 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

T: Fachverband Teilchenphysik

T 71: Data analysis, Information technology III

T 71.2: Talk

Wednesday, March 17, 2021, 16:15–16:30, Tu

Studies of modern machine learning methods for tau lepton identification with the CMS detector — •Andrew Issac¹, Günter Quast¹, Roger Wolf¹, Stefan Wunsch^1,2, and Sebastian Brommer¹ — ¹Karlsruhe Institute of Technology, Institute of Experimental Particle Physics, Karlsruhe, Germany — ²CERN, Geneva, Switzerland

High-energy physics particle reconstruction algorithms are a perfect playground for modern machine learning methods due to the vast amount of data and complex learning tasks. The tau lepton identification is an interesting field to explore new neural network models such as graph-based networks and benchmark them against existing approaches, e.g., a deep convolutional model currently used by CMS.