Speaker
Krunal Bipin Gedia
(ETH Zurich (CH))
Description
Hadronic jets coming from the fragmentation of b-quarks are crucial tools for a number of physics channels at the CERN LHC, ranging from the Higgs physics to searches for physics beyond the Standard Model. We present a technique that allows tuning the simulated response of the CMS detector at the LHC to b-jets. Machine learning algorithms and likelihood fits are used to obtain finely-grained correction factors, based on samples of b-jets from ttbar decays. Specifically, we employ multivariate classifiers and non-linear multi-dimensional quantile regression models to tune the detector response to b-jets with different properties and compositions.
Primary author
Krunal Bipin Gedia
(ETH Zurich (CH))
