This dissertation presents a search for τ ̃ pair production conducted with 137 fb−1 of data collected between 2016 and 2018 by the Compact Muon Solenoid (CMS) experiment at the Large Hadron Collider (LHC). Supersymmetry (SUSY) can provide an elegant solution to the Higgs mass hierarchy problem, and supersymmetric models with the τ ̃ lepton as the lightest supersymmetric particle can potentially explain the dark matter relic density via τ ̃ and lightest SUSY particle annihilation. In this search we look for purely left-handed and purely right-handed τ ̃ pair production, as well as the degenerate scenario where left- and right-handed pairs are produced together. A result with long- lived τ ̃s that give rise to hadronically decaying τ leptons is presented for the first time. The search is performed in the all-hadronic final state, and the background estimation uses data-driven and hybrid data-driven and simulation techniques. The observed yields are consistent with Standard Model predictions. Masses below 350 GeV and below 400 GeV are excluded at the 95% confidence level in the left-handed and degenerate scenario respectively.

Standard model (SM) four-top quark production, where proton-proton collisions produce two top-antitop pairs, is a rare process with great potential to reveal new physics. Measurement of the cross section is not only a direct probe of the top quark Yukawa coupling with the Higgs, but an enhancement of this cross section is predicted by several beyond the standard model (BSM) theories. This process is studied in fully-hadronic proton-proton collision events collected during Run II of the CERN LHC by the CMS detector, which corresponded to an integrated luminosity of 137 fb^−1 and a center of mass energy of 13 TeV. In order to optimize signal sensitivity with respect to significant and challenging backgrounds, several novel machine-learning based tools are applied in a multi-step and data-driven approach. Boosted decision tree (BDT) and deep neural net (DNN) based hadronic top taggers are used to identify hadronically decaying top quark candidates with moderate and high transverse momenta, respectively, in order to suppress backgrounds and categorize events by the multiplicity of reconstructed top tags, and an event-level kinematic BDT distribution is subsequently used to extract the signal. Control regions inspired by the “ABCD” method are used to obtain a data-driven estimate of the background, and data distributions in these control regions are given as inputs to a DNN in order to estimate the event-level BDT discriminant distributions of the major backgrounds. The observed signal strength, defined as the ratio of the observed four-top production rate to the standard model expectation, is measured to be 5.1^{+2.3}_{-2.0}. The corresponding observed (expected) significance and limit times the SM cross section are 2.25 (0.43) standard deviations and 8.39 (4.88) respectively. A combination of this result with multiple final states is in progress as of the writing of this thesis, and future BSM interpretations to investigate this larger-than-expected signal strength are planned.

The dissertation herein presents the latest public search for direct tau slepton (τ ̃), the superpartner of the tau lepton) production at the Large Hadron Collider (LHC). The search is performed using data collected from the Compact Muon Solenoid (CMS) experiment. A total of 77.2 fb−1 of data from proton-proton (pp) collisions at 13 TeV are recorded and analyzed. The analysis probes the final state with the largest branching fraction from τ ̃ decays, that of two hadronic taus (τ).

No significant deviation from Standard Model expectations are observed in the data and accordingly, a 95% confidence level upper limit on the direct stau pair production cross section is set. Mass degenerate stau pair production is excluded for minimally supersymmetric models (MSSM) with m(τ ̃) ∈ 90-150 GeV. Similarly, exclusion is expected for left-handed only pair production with m(τ ̃)=125 GeV. However, no such exclusion is observed in the data.

These results motivate an additional study carried out in a simulation of the High-Luminosity LHC (HL-LHC) environment with the CMS Phase-2 detector. This scenario corresponds to 3000 fb−1 of data taken at a center-of-mass energy of 14 TeV. The study shows sufficient sensitivity to formally expect the discovery of mass degenerate direct stau pair production for m(τ ̃) ∈ [100, 650 GeV].

Measurement of the decay of the Higgs boson to charm quarks provides a direct probe of the Higgs coupling to second-generation quarks. Therefore, it is crucial for understanding the structure of Yukawa couplings. In this thesis, a search for the Higgs boson decaying to charm quarks with the CMS experiment is presented. The search is designed for Lorentz-boosted Higgs bosons produced in association with vector (V) bosons (W or Z bosons). A novel approach that reconstructs both quarks from the Higgs boson decay with a single large-radius jet is adopted. The charm quark pair is identified with an advanced deep learning--based algorithm. This approach leads to a highly competitive result: Using proton-proton collision data corresponding to an integrated luminosity of 35.9 fb-1, an observed (expected) upper limit on the VH cross section times the H->cc branching fraction of 71 (49) times the standard model expectation at 95% confidence level is obtained.

A detailed description of the deep learning--based boosted object identification algorithm is also presented in this thesis. It is a versatile algorithm designed to identify and classify hadronic decays of highly Lorentz-boosted top quarks and W, Z, Higgs bosons. Using deep neural networks to directly access and process the raw information of all constituent particle-flow candidates of a jet, this advanced algorithm has achieved significant performance improvements compared to traditional approaches.

Higgs decay to a pair of charm quarks is a process that has potential to shed light on the standard model Higgs-charm Yukawa coupling and BSM physics in the form of two Higgs doublet models (2HDMs). However, it has not yet been observed experimentally. TThis thesis presents the current status of a search for the titular t$\bar{\text{t}}$H(h$\rightarrow$ c$\bar{\text{c}})$ process. The primary focus in this work is the single-lepton channel, which will utilize the full CMS Run II data set of 138 fb$^{-1}$. This analysis takes advantage of recent advances in machine learning to obtain high jet-tagging and event classification performance, and reports a preliminary new constraint on the t$\bar{\text{t}}$H(h$\rightarrow$ c$\bar{\text{c}})$ cross-section.

As successful as the Standard Model of particle physics has been it still has several major shortcomings which range from unanswered theoretical questions to a lack of any explanation for observed phenomena such as dark matter. One proposed theory for physics beyond the Standard Model which provides solutions for some of these issues is supersymmetry. This dissertation presents a search for supersymmetry using 2.3 fb^(-1) of proton-proton collision data. This data was collected at a center-of-mass energy of 13 TeV by the CMS detector at the LHC during 2015. This search focuses on top squark pair production where the produced stops both decay to an all hadronic final state. These decays are characterized by multiple jets and missing transverse momentum. A baseline search region is defined to be sensitive to signal processes which occur at rates many orders of magnitude lower than Standard Model processes. The sensitivity to various signal models is improved by dividing this baseline region into distinct categories. Events with an unreconstructed lepton from leptonic W boson decays constitute the primary background. There are also significant contributions from events where a Z boson decays invisibly to neutrinos especially in bins with higher missing transverse momentum. Events with multijet production where one jet has been severely mis-measured as well as those with a pair of top quarks and an invisibly decaying Z boson also have a small presence in the search region. The contributions from these processes to the search region is estimated using data control regions. No statistically significant deviations from the predicted background yields are observed. The results are interpreted in terms of exclusion limits using the Simplified Model Spectrum framework. Stop and neutralino masses are probed up to 780 GeV and 260 GeV respectively.