Master Thesis Open Access
Daigler, Simon
Klute, Markus; Wolf, Roger
A search for non-resonant Higgs-boson pair production (HH) in the bbππfinal state is presented at √π = 13 TeV using the 2018 CMS proton–proton dataset corresponding to 59.8 fb−1 after a common re-optimization of the reconstruction and calibration of the data by the CMS Collaboration. The resolved topology is targeted and three π+π− final states πhπh, ππh and eπh are taken into account, where πh refers to the decay of the π lepton into hadrons. Object reconstruction and identification are based on Particle Flow, DeepJet and DeepTau v2.5 algorithms. A multi-class classification neural network is used for event discrimination and category definition. The background from QCD multijet events is estimated from data via an ABCD method using the charges and DeepTau vsJet discriminants to define corresponding control regions. The di-π invariant mass, a key observable for discriminating signal from background, is reconstructed using the FastMTT algorithm.
Expected upper limits are derived from a profile-likelihood fit to the Asimov dataset. The median expected 95 % CL upper limit on the signal strength π= π(pp →HH)/πSM is found to be πexp95 = 10.9, about 2 % weaker than the result obtained when considering only statistical uncertainties (πexp95 = 10.7). Projections obtained from scaling rates with integrated luminosity while keeping shapes and the size of nominal systematics yield median expected 95 % CL limits of πexp95 = 7.2 for the full Run 2 corresponding to an integrated luminosity of 138 fb−1 and πexp95 = 4.7 for a Run 2 + Run 3 scenario with 500 fb−1.
| Name | Size | |
|---|---|---|
|
Masterthesis_SimonDaigler.pdf
md5:fafd59f1b06a0c9174ce67c6a773be48 |
13.8 MB | Download |