°µÍø½ûÇø

I am active in a variety of research projects with the CMS experiment spanning data analysis searching for physics beyond the standard model to developing cutting edge technology for real-time physics reconstruction algorithms for the CMS Level 1 trigger. Some current projects in our group are outlined below.

Current Projects

• Anomaly Detection Triggers We are developing a new method to select LHC collisions to record for further study. A machine learning algorithm known as a variational autoencoder is trained on CMS data to select the most anomalous events. This unsupervised approach delivers a dataset without any bias toward a specific new physics model. Related publications: , .
• Model Independent Searches with Anomaly Detection Our group helped implement the first anomaly detection trigger running at the hardware (FPGA) level, which we call AXOL1TL. We are now working to analyze this unique dataset to search for new physics in a model independent way in the events that AXOL1TL characterized as anomalous.
• Searches for Lepton Flavor Violation Lepton flavor universality holds that electrons, muons, and taus will all have identical couplings to other standard model particles. We are testing this fundamental assumption in precision measurements of B meson decays (Rk ratio) and in searches for rare decays of Z bosons to different flavor leptons. Related publications:
• Searches for Supersymmetry - Higgsinos Supersymmetry posits that all known fundamental particles have supersymmetric partners, which may hold the key to understanding electroweak symmetry breaking and explain dark matter. We are searching for the partner of the Higgs boson, known as the higgsino in a variety of final states. Related publications: , , , , , , .
• CMS Trigger Upgrades - Track Trigger Charged particle tracking is the most combinatorially challenging step in the CMS event reconstruction. Until now, this has prevented CMS from including tracking in the Level 1 Trigger reconstruction. We are working on upgrades to the L1T to reconstruct tracks at the full 40 MHz LHC collision rate in algorithms implemented in high speed FPGAs. Related publications: , , .
• CMS Trigger Upgrades - Global Track Trigger The GTT is a sub-system in the CMS Level 1 Trigger that receives all of the charged tracks in an event. We are developing algorithms in firmware to use these tracks to reconstruct physics objects including the primary vertex, missing energy, displaced jets, and W to 3 pion candidates.
• US-CMS Trigger/DAQ Leadership Since 2018, I have served as co-coordinator for the US-CMS Level 1 Trigger upgrade project. This includes overseeing L1T scope from both the Department of Energy and the National Science Foundation. Related publications: .
• Searches for Emerging Jets Dark matter is well known to be abundant in the universe from astronomical observations of its gravitational pull. However, no experiments thus far have been able to directly observe dark matter interactions. We are searching for a class of dark matter models that interact with the standard model through small couplings to a dark QCD-like sector. Related publications: , .
• Smart Pixels Embedding machine learning algorithms directly in the readout electronics for silicon tracking detectors has the potential to allow these "smart" sensors to provide physics information at the detector level before reconstructing any full particle trajectories. We are investigating the physics potential of this technology for future detectors. Related publications: .