Speaker
Description
The rare Higgs boson decay to two muons provides the best opportunity to measure the Higgs boson's coupling to a second generation fermion. The ATLAS collaboration at CERN has recently established evidence for this decay at 3.4 standard deviations ($\sigma$) using data from Run 2 and part of Run 3. Significance is expected to increase as the remaining Run 3 data from the Large Hadron Collider (LHC) is collected and analyzed, but it is not expected to meet the stringent 5.0 $\sigma$ requirement to announce a discovery. Estimates place discovery during the High-Luminosity LHC (HL-LHC) era during which seven times as much data as Runs 2 + 3 combined will be collected. Standard model processes like $H\to\mu\mu$ can be simulated ahead of time using the expected conditions of the future HL-LHC to provide estimates of the experimental reach of such rare physics processes. In particular, upgrades to the ATLAS detector for the HL-LHC era are expected to provide higher-quality information that can be used for classifying $H\to\mu\mu$ events. The analysis of $H\to\mu\mu$ during Run 2 + 3 has so far relied heavily on boosted decision trees to provide background reduction in the signal region, but studies suggest high-information environments could benefit from using neural network discriminators instead. This talk investigates the use of a neural network discriminator in the search for $H\to\mu\mu$ and discovery prospects using simulated HL-LHC data.
| Keyword-1 | Higgs |
|---|---|
| Keyword-2 | Neural Nets |
| Keyword-3 | HL-LHC |