Machine learning approach towards an improved vector boson fusion selection

The strength of the Higgs boson coupling to other fermions and bosons is exhaustively studied during the first two runs of the LHC. The interacting with bosons, third generation fermions and muons agree well with the theoretical predictions up to now. The range of investigated coupling partners needs to be extended to also include the charm quark. The main challenge hereby is to differentiate between background from QCD processes and signal events. A promising Higgs production mode to observe the coupling is the vector boson fusion Higgs production due to its large cross section and promising separability from background. In this project the current cut-based selection on the VBF Higgs production is verified and an improved approach is introduced by developing three different machine learning algorithms. These machine learning algorithms become optimized and yield substantial improvements compared to the cut-based selection.