Boosted Decision Tree Optimization for the ATLAS search of ttH production in the 2l same-sign channel

The main goal is to have a direct measurement of the Yukawa coupling of the Higgs boson to the top quark which is only possible in the production process 𝑝𝑝 → ttH + 𝑋. In this analysis, final states with 2 same sign leptons (neutrinos not counted) have been used in order to estimate the expected significance of the ttH process. A study using Boosted Decision Trees was done using Monte Carlo simulation equivalent to a luminosity of 36.5 fb-1 at √s=13 TeV, characteristics of the years 2015 and 2016 of Run-2 at LHC. The focus of my summer student program was to investigate the performance of the BDT, mainly: To avoid building of a rigid and possible overtrained BTD (Boosted Decision Tree) in charge of identifying pp→ttH+X process by removing systematically the number of variables used in the analysis. Look the expected sensitivity's dependence on different parameters that takes in account the BDT.