April 23, 2023
The American Physical Society (APS) hosted the annual April meeting, “Quarks to Cosmos.” The meeting encompassed a broad range of physics: astrophysics, particle physics, nuclear physics, and gravitation. Researchers from around the world represented more than twenty APS Units and committees.
These 2nd-year students were participants in the Particle Physics SIR Course under the direction of physics faculty member Dr. Peter Dong. George Bayliss and Jesus Fileto were also invited to give talks on their projects and present their posters with their classmates. All abstracts are in the IMSAloquium 2023 event booklet on Digital Commons.
Gautham Anne’s project titled Automatic Datacard Generation and Significance Estimation with Punzi Criterion for Higgs Analyses showcased how a framework was developed to incorporate into their analysis group’s framework to automatically generate datacards by using the Higgs Combine Tool to test combining multiple generated datacards.
George Baylis’ project titled A Comparison of Various Dark Photon Production Mechanisms presented five different production mechanisms, or portals, which generated dark photons from Higgs bosons, supersymmetric particles, and Z’ bosons. Event signatures were analyzed, showing that limits for each could be parameterized from a single portal.
Surya Bhamidi’s project, Parametrizing Doubly Charged Higgs Invariant Mass Histograms, focused on finding a peak for any mass point, not only the specific mass points already obtained from the Monte Carlo data. Many functions obtained from the fits were parameterized in terms of mass, and the parameterized functions were then used to perform an unbinned mass likelihood fit.
Dean Cianciolo’s project titled Automatic Histogram Generation for Multi-Channel Analyses investigated how the Monte Carlo analysis validated that the more precise Monte Carlo estimates were not far off previous generations and generated 36 different plots in less than 5 minutes from a single script.
Jesus Fileto’s project, An Investigation of Lepton Jet Kinematics, Fakes, and Production from Dark Photons, showed that the Higgs production mechanism produced two dark photons, resulting in narrow lepton jets arising from these dark photons. Monte Carlo data was then analyzed to find distinctive signatures of these lepton jets, such as transverse momenta and eta, which define their origin from dark photons.
Kevin Huang’s project titled, Integration of CMSSW software into the Analysis Framework, showed that the analyzer needed to be restructured to process events individually, as the event data module (EDM) format was structured to be passed into the analyzer as individual events rather than a root file. A cmsRun command was established to avoid modification of the analytical framework.
Rohan Jain’s project titled, Optimizing Trigger Selection for Detection of Doubly Charged Higgs Bosons at the LHC, aimed to programmatically find the most efficient triggers for selecting H++ events for application in the Compact Muon Solenoid experiment and produce an outcome for creating a new figure of merit.
Caroline Kowal’s project, Utilizing CRAB for Limit Findings, focused on obtaining faster results by simulating specific events through CRAB so jobs could run simultaneously and utilized outputs by finding data limits, which assisted in particle events analysis.
Jack Morby’s project titled, The Effectiveness of a Localized b-Jet Veto for Improving Lepton Jet Reconstruction, eliminated a source of background in the search for the dark photon and further analyzed the extent to which efficiency changes were compiled. The framework was updated with an “Info Map” to support adding variables utilized during analysis.
James Pan’s project titled, Drell-Yan Background for Doubly Charged Higgs, investigated how the H++ and H– can each decay to a pair of like-sign leptons (muons or electrons in this study) with TeV-scale mass, providing a distinctive event signature. Causes of Drell-Yan background were determined, and a background estimate was then constructed, which led to identifying underlying factors of each cause to use other measurements to predict their rate.
Dheeran Wiggins’ project titled, Unidirectional Build Architecture: Refactoring a HEP Data Analysis Codebase, explored an alternative architecture for the H±± and γd searches through codebase refactoring, reorganizing the grouping and placement of methods, classes, and files. The data collection directory was then refactored into seven task-specific directories arranged with hierarchical directives to prevent circular dependencies.
Kevin Zhang’s project, Multivariate analysis for detecting lepton jets, showed that the leading particle transverse momentum and the difference in transverse momentum between the leading and runner-up particles increased model effectiveness. The resulting ROC curves, plots, input parameters, and the distribution of model prediction on testing data were presented.