Wing Sheung Chan

58 Object reconstruction and identification 0 0.2 0.4 0.6 0.8 1 e -veto BDT output 10 2 10 3 10 4 10 5 10 6 10 7 10 Number of t had-vis candidates [A.U.] True t lepton (train) Electrons (train) True t lepton (test) Electrons (test) (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Efficiency 1 10 2 10 3 10 4 10 5 10 Rejection | h | < 1.37 Deprecated e -veto BDT Likelihood-based veto Redeveloped e -veto BDT (b) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Efficiency 1 10 2 10 3 10 4 10 5 10 Rejection Deprecated e -veto BDT Likelihood-based veto Redeveloped e -veto BDT 1.37 < | h | < 2.0 (c) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Efficiency 1 10 2 10 3 10 4 10 5 10 Rejection | h | > 2.0 Deprecated e -veto BDT Likelihood-based veto Redeveloped e -veto BDT (d) Figure 3.5.: The (a) expected output distributions and (b)– (d) the ROC curves of the e -veto BDT. The output distributions evaluated for both the training and testing samples are shown. The ROC curves for τ had - vis candidates in different | η | regions are shown, alongside the ROC curves of the deprecated e -veto BDT used during Run 1 and an alternative likelihood-based vetoing method used in the beginning of Run 2.