Andrew Brinkerhoff

The first implementation of a Machine Learning Algorithm inside a Level-1 trigger system at the LHC is presented. The Endcap Muon Track Finder (EMTF) at CMS uses Boosted Decision Trees (BDTs) to infer the momentum of muons in the forward region of the detector, based on 25 different variables. Combinations of these variables representing 230 distinct patterns are evaluated offline using regression BDTs. The predictions for the 230 input variable combinations are stored in a 1.2 GB look-up table in the EMTF hardware. The BDTs take advantage of complex correlations between variables, the inhomogeneous magnetic field, and non-linear effects – like inelastic scattering – to distinguish high momentum signal muons from the overwhelming low-momentum background. The new momentum algorithm reduced the background rate by a factor of three with respect to the previous analytic algorithm, with further improvements foreseen in the coming year.

The performance of electromagnetic calorimeter modules made of proton-irradiated PbWO4 crystals has been studied in beam tests. The modules, similar to those used in the Endcaps of the CMS electromagnetic calorimeter (ECAL), were formed from 5×5 matrices of PbWO4 crystals, which had previously been exposed to 24 GeV protons up to integrated fluences between 2.1× 1013 and 1.3× 1014 cm−2. These correspond to the predicted charged-hadron fluences in the ECAL Endcaps at pseudorapidity η = 2.6 after about 500 fb−1 and 3000 fb−1 respectively, corresponding to the end of the LHC and High Luminosity LHC operation periods. The irradiated crystals have a lower light transmission for wavelengths corresponding to the scintillation light, and a correspondingly reduced light output. A comparison with four crystals irradiated in situ in CMS showed no significant rate dependence of hadron-induced damage. A degradation of the energy resolution and a non-linear response to electron showers are observed in damaged crystals. Direct measurements of the light output from the crystals show the amplitude decreasing and pulse becoming faster as the fluence increases. The latter is interpreted, through comparison with simulation, as a side-effect of the degradation in light transmission. The experimental results obtained can be used to estimate the long term performance of the CMS ECAL.

With the accumulation of large collision datasets at a center-of-mass energy of 13 TeV, the LHC experiments can search for rare processes, where the extraction of signal events from the copious Standard Model backgrounds poses an enormous challenge. Multivariate techniques promise to achieve the best sensitivities by isolating events with higher signal-to-background ratios. Using the search for Higgs bosons decaying to two muons in the CMS experiment as an example, we describe the use of Boosted Decision Trees coupled with automated categorization for optimal event classification, bringing an increase in sensitivity equivalent to 50% more data.

We present a search for the standard model Higgs boson produced in association with a top quark pair in 5 fb−1 of 7 TeV pp collision data. We look for events where the Higgs boson decays to b and the top quark pair decays to either lepton plus jets (t → ℓνjjb) or dileptons (t → ℓνℓνb), where ℓ is an electron or muon. The major background to our signal is top pair production. We use artificial neural networks to discriminate between background and signal events. We perform a simultaneous fit for signal and background fractions using the neural network output distributions. Our expected limit on Higgs boson production cross section times H → b branching ratio for a Higgs boson mass of 125 GeV/c2 is 4.6 times the standard model expectation, while the observed limit is 3.8 times the standard model expectation.

The first implementation of Boosted Decision Trees (BDTs) inside a Level-1 trigger system at the LHC is presented.The Endcap Muon Track Finder (EMTF) at CMS uses BDTs to infer the momentum of muons in the forward region of the detector, based on 25 different variables.Combinations of these variables are evaluated offline using regression BDTs, whose output is stored in 1.2 GB look-up tables (LUTs) in the EMTF hardware.These BDTs take advantage of complex correlations between variables, the inhomogeneous magnetic field, and non-linear effects such as inelastic scattering to distinguish high-momentum signal muons from the overwhelming low-momentum background.The LUTs are used to turn the complex BDT evaluation into a simple look-up operation in fixed low latency.The new momentum assignment algorithm has reduced the trigger rate by a factor of 3 at the 25 GeV trigger threshold with respect to the legacy system, with further improvements foreseen in the coming year.