Calibration of the Boosted $X\rightarrow b\bar{b}$ Tagger in $g\rightarrow b\bar{b}$ splitting events at the ATLAS Experiment

The physics programme at ATLAS involves a variety of Standard Model and Beyond Standard Model resonances decaying to two b quarks, including the Higgs Boson. In order to identify the decays of these particles at high momentum, ATLAS has developed the Boosted $X\rightarrow b\bar{b}$ tagger, which needs to be calibrated. The double b-tagging efficiency scale-factors between MC and data are derived using boosted $g\rightarrow b\bar{b}$ splitting in multi-jet events. MC simulations are used in the estimates of the fraction of $b\bar{b}$ events in data. However, simulation of QCD is notoriously imperfect, making it necessary to use a data-driven approach to correct the flavour composition of the MC, such that the double b-tagging scale factors can be derived. This is done by fitting the signed transverse impact parameter distributions of sub-jet tracks in data and rescaling the MC accordingly. This poster presents the calibration of boosted $b\bar{b}$ tagging using 36.1 fb$^{-1}$ of ATLAS Run 2 pp collision data gathered in 2015 and 2016. A brief overview of the techniques employed by the boosted $X\rightarrow b\bar{b}$ tagger and relevant event selections are given. The procedure to correct the flavour composition of simulated QCD samples is described, and current calibration scale factors are presented. Ongoing and future efforts are summarised.