The study of rare processes is a way to probe the existence of physics beyond the Standard Model (SM). In the Standard Model with massless neutrinos, the three lepton flavor numbers are exactly conserved. The observation of neutrino oscillations anyway provides a mechanism, through neutrino loops, for lepton flavor violating decays of charged leptons such as tau->3mu. While in the SM this process has an extraordinarily small branching fraction, several SM extensions predict a much larger branching fraction, with values accessible to current experiments.
This thesis aims at searching for the tau->3mu decay using the data collected by the CMS experiment at LHC, exploiting tau production mechanisms never considered before for this search. The student will join the CMS team involved in this search. He/she will perform a real data analysis, which includes the development and implementation of advanced Machine Learning techniques to boost the search sensitivity.
Keywords: CMS, Rare decays, Data Analysis, Machine Learning
The discovery of a new particle, in the mass region around 125 GeV consistent with the Higgs boson predicted by the Standard Model, was announced by the CMS experiment on the 4th July in 2012 at the Large Hadron Collider.
Two of the main discovery channels were H→ZZ→4 leptons and H→gammagamma. They are defined as golden channels thanks to the complete reconstructed final state and very good momentum resolution. These processes have been deeply used to study Higgs boson properties looking for possible contributions beyond the Standard Model.
The student, profiting of these channels, will perform real data analysis, which includes the development and implementation of advanced Machine Learning techniques to measure these properties and squeeze the analysis sensitivity, using data collected by CMS detector during not only Run 2 but also Run 3, at the new LHC energy of 13.6 TeV.
Keywords: CMS, Higgs Boson, Data Analysis, Machine Learning
Heavy-ion collisions at the LHC recreate droplets of the matter that filled the universe a microsecond after the Big Bang. This matter, called the Quark Gluon Plasma (QGP), too hot to be formed by protons or neutrons or any other hadrons, behaves like a perfect liquid.
We are interested in understanding the properties of this new state of matter and for that we use jets. Jets[2] - the spray of hadrons produced in the fragmentation and subsequent hadronization of an energetic quark or gluon - are ideal probes because they interact with the QGP at different scales, and this is imprinted in their internal structure.
With this thesis, we will first gain insight on newly proposed jet energy correlators[3,4] by implementing them in Montecarlo simulations and studying their big potential sensitivity to quark mass and to QGP-induced signal. This will be followed by a feasibility study of the actual measurement in CMS, evaluating detector effects and the smearing of the large uncorrelated background that is present in heavy-ion collisions. A matching criteria for the correlators will be designed and a multidimensional Bayesian unfolding procedure will be set up and exercised on real data. The student will be trained to perform these tasks at each step of the project
Keywords: CMS, Heavy Ion Physics, Quantum Chromodynamics
Theoretical scenarios with new long-lived particles are of great interest in direct searches for physics beyond the Standard Model at LHC. To detect unusual signatures of massive long-lived particles, non-standard algorithms and experimental techniques need to be developed. For the high luminosity phase of LHC (HL-LHC), the CMS experiment will build a new Mip Timing Detector (MTD) aimed at measuring the time-of-flight of charged particles with an excellent target resolution of 30 ps. This precise time information can be used to detect signals of new long-lived particles that travel inside the detector for a certain distance before decaying to photons, leptons or jets.
The student will develop identification algorithms to detect these long-lived particles using the novel time information from MTD and study the impact of these methods on the sensitivity to new physics processes. This research activity will contribute to increase the new physics discovery potential of the CMS experiment at HL-LHC.
Keywords: CMS, Exotic New Physics Signatures, Precise Timing, Multivariate Analysis Techniques
Intriguing scenarios of new physics provide explanation to several shortcomings of the Standard Model, as for example the source of parity violation in the weak sector, matter dominance of the universe or neutrino masses.
Such models usually lead to nonconventional signatures at the LHC. The latter are varied and, by nature, often very different from signals of Standard Model processes. They could include new particles with unusual properties as for example fractionally charged particles, long-lived particles decaying in the outer regions of the CMS silicon tracker or in the calorimeters or hypothetical heavy states, for instance the heavy Majorana neutrinos.
Novel experimental techniques, including creative usage of CMS sub detectors and advanced machine learning techniques can be employed to identify such signals. The huge amount of data collected by CMS during the LHC Run 2 and the very fresh Run 3 data, offer the unique opportunity to develop novel techniques.
In this thesis the student will join the efforts of an analysis team working on one of such peculiar signatures. He/She will develop unique analysis strategy and tools that could enhance the sensitivity of the search and could boost its discovery reach.
Keywords: CMS, Exotic New Physics Signatures, Machine Learning Techniques
Future LHC running periods will be characterized by very hard experimental conditions, reaching in the high-luminosity phase (HL-LHC) ~200 simultaneous (pile-up) interactions per bunch-crossing. For this reason, CMS has proposed for its upgrade program a novel timing detector for charged particles (MTD) with an expected time resolution of 30ps. Time information will allow to significantly reduce the pile-up contamination, with large improvements for the event reconstruction. This upgrade will be equivalent to an increase of 20-30% of the integrated luminosity collected at the HL-LHC, with clear benefits for the physics reach of the CMS experiment.
Time informations from the new Mip Timing detector are combined with the data from inner spatial tracking detectors in order to reach a 4-dimensional event reconstruction. In this thesis, the student will develop novel and innovative machine learning algorithms that will allow to fully exploit the potential this new timing detector.
Keywords: CMS, Event Reconstruction, Machine Learning, Precise Timing Detector