Powered by Indico
v3.2.3
Abstract:
Identification of highly Lorentz-boosted particles, such as the top quark and the W, Z, and Higgs boson, provides powerful handles for standard model measurements and new physics searches at the LHC. In this talk, I present several new approaches for boosted jet tagging with machine learning (ML), highlighting the importance of symmetry and physics principles on the design of ML taggers. These new techniques have brought significant performance improvement, as demonstrated in a number of recent CMS physics analyses.
Biography:
Huilin Qu obtained his BSc in Physics from Peking University in 2014 and his PhD in experimental particle physics from University of California, Santa Barbara in 2019. He was a postdoctoral researcher at UC Santa Barbara from 2019 to 2020, and is currently a Senior Research Fellow at CERN. Huilin’s research has focused on searches for new physics and measurements of the Higgs boson properties at the LHC, particularly with novel approaches and new techniques such as machine learning. He developed the first search for the Higgs boson decaying to a pair of charm quarks with the CMS experiment, and led the search for supersymmetric partners of the top quark in the all-hadronic final state. In addition, Huilin proposed several novel machine learning-based approaches for boosted jet tagging, including the DeepAK8 and the ParticleNet algorithms, which substantially improved the performance and have become the standard in CMS for relevant physics analyses.
Online link: https://meeting.tencent.com/s/i19M98QqDqla (ID: 747804504 Password: 123456）)