Research in our group
CMS experiment at the CERN LHC collects millions of events with top quarks every year. With such a large sample, we can understand the properties of the top quark as well as probe some of the more rare processes with top quarks.
Measurement of the tt-bar + bb-bar process - this is one of the backgrounds to tt-bar + Higgs process
Measurement of the differential cross section in tt-bar events
Search for the production of four top quarks
CP Violation in top quark
Analysis techniques - explore application of various advanced analysis techniques to problems in particle physics
Deep learning - Jet reconstruction, complex event classification, using generative methods to predict background data
Recent research with students as primary authors
Improved extrapolation methods of data-driven background estimations in high energy physics
European Physical Journal 81, Article number: 643 (2021).
New method (Extended ABCD method) of background estimation shows large improvement compared to the standard method (ABCD method) .
Measurement of the CKM matrix element |V_cb| from top quark pair events at LHC
Journal of Korean Physical Society 78, 1023–1028 (2021)
A new direct way of measuring CKM matrix element |V_cb| free from hadronization uncertainty is shown. Within a few years, it should be measurable at the LHC.
A data-driven event generator for Hadron Colliders using Wasserstein Generative Adversarial Network
Journal of Korean Physical Society 78, 482-489 (2021)
We explored using deep neural network called GAN to generate realistic but fake data. We showed that GAN is able to find multidimensional distribution and generate genuinely new events that were not present in the training data.
