[5] Summary of several Standard Model total and fiducial production cross section measurements, corrected for leptonic branching fractions, compared to the corresponding theoretical expectations. All theoretical expectations were calculated at NLO or higher. The dark-color error bar represents the statistical uncertainty. The lighter-color error bar represents the full uncertainty, including systematics and luminosity uncertainties. The data/theory ratio, luminosity used and reference for each measurement are also shown. Uncertainties for the theoretical predictions are quoted from the original ATLAS papers. They were not always evaluated using the same prescriptions for PDFs and scales. The Wgamma and Zgamma theoretical cross-sections have non-perturbative corrections applied to the NNLO fixed order calculations [6]. Not all measurements are statistically significant yet.

[13] A diagram of the Standard Model of particles. Shown are three generations of twelve fermions (quarks and leptons), all with spin frac12. The five force carriers (bosons) are shown: gluon, photon, W/Z bosons, and the Higgs boson. Also depicated is the graviton, a theoretical mediator of the gravitational force which is not currently in the Standard Model. All gauge bosons, except for the Higgs boson.

The elementary processes of Quantum Electrodynamics. Note that time is horizontal (a convention in ATLAS). In (a), a charged particle, f, enters, emits (or absorbs) a photon, γ, and exits. In (b), the Z boson mediates such processes. In (c), a lepton converts into corresponding neutrino with emission or absorption of W^±. These diagrams were made with TikZ-Feynman [38].

The elementary processes of Quantum Electrodynamics. Note that time is horizontal (a convention in ATLAS). In (a), a charged particle, f, enters, emits (or absorbs) a photon, γ, and exits. In (b), the Z boson mediates such processes. In (c), a lepton converts into corresponding neutrino with emission or absorption of W^±. These diagrams were made with TikZ-Feynman [38].

The elementary processes of Quantum Electrodynamics. Note that time is horizontal (a convention in ATLAS). In (a), a charged particle, f, enters, emits (or absorbs) a photon, γ, and exits. In (b), the Z boson mediates such processes. In (c), a lepton converts into corresponding neutrino with emission or absorption of W^±. These diagrams were made with TikZ-Feynman [38].

A few Feynman diagrams of basic QCD interactions to lowest order, including (a) gluon radiation, (b) quark/anti-quark annihilation, (c) gluon splitting, and (d) gluon self-coupling. These diagrams were made with TikZ-Feynman [38].

A few Feynman diagrams of basic QCD interactions to lowest order, including (a) gluon radiation, (b) quark/anti-quark annihilation, (c) gluon splitting, and (d) gluon self-coupling. These diagrams were made with TikZ-Feynman [38].

A few Feynman diagrams of basic QCD interactions to lowest order, including (a) gluon radiation, (b) quark/anti-quark annihilation, (c) gluon splitting, and (d) gluon self-coupling. These diagrams were made with TikZ-Feynman [38].

A few Feynman diagrams of basic QCD interactions to lowest order, including (a) gluon radiation, (b) quark/anti-quark annihilation, (c) gluon splitting, and (d) gluon self-coupling. These diagrams were made with TikZ-Feynman [38].

A Feynman diagram at LO with probability amplitude proportional to the square of the strong coupling constant. This particular Feynman diagram represents the interaction between quarks that, for example, binds them into hadrons. These diagrams were made with TikZ-Feynman [38].

[48] Cartoon of the (a) cluster hadronization model which treats individual color singlets separately and the (b) lund string hadronization model which propagates field lines of color flux. These two leading models approximate the non-perturbative process of hadronization to map colored partons onto stable, colorless hadrons.

[48] Cartoon of the (a) cluster hadronization model which treats individual color singlets separately and the (b) lund string hadronization model which propagates field lines of color flux. These two leading models approximate the non-perturbative process of hadronization to map colored partons onto stable, colorless hadrons.

[49] Sketch of a hadron-hadron collision as simulated by a Monte-Carlo event generator. The red blob in the center represents the hard collision, surrounded by a tree-like structure representing Bremsstrahlung as simulated by parton showers. The purple blob indicates a secondary hard scattering event. Parton-to-hadron transitions are represented by light green blobs, dark green blobs indicate hadron decays, while yellow lines signal soft photon radiation.

[52] MMHT2014 NNLO PDFs at Q² = SI10GeVsquared and Q² = 10⁴ siGeVsquared with associated 68% confidence-level uncertainty bands. The function x f(x, Q²) is plotted versus x for different flavors: u, ubar, d, dbar, s=sbar, and g.

[52] MMHT2014 NNLO PDFs at Q² = SI10GeVsquared and Q² = 10⁴ siGeVsquared with associated 68% confidence-level uncertainty bands. The function x f(x, Q²) is plotted versus x for different flavors: u, ubar, d, dbar, s=sbar, and g.

Feynman diagrams showing the top quark decays for (a) hadronic and (b) leptonic. These diagrams were made with TikZ-Feynman [38].

Feynman diagrams showing the top quark decays for (a) hadronic and (b) leptonic. These diagrams were made with TikZ-Feynman [38].

[4] An example of a loop diagram which corrects the Higgs mass.

[4] An updated version of creffig:susy:loopcorrection with the inclusion of the stop squark, top quark one-loop diagrams. The stop squark is a bosonic superpartner of the fermionic top quark, provides equal and opposite contribution to the top quark loop, cancelling out the contribution. There are two loops because there are two bosonic partners for the top quark, a fermion with spin.

[92,93] Theoretical cross-sections of gluino pair production are shown in the model of equal degenerate squark masses, as a function of gluino mass at √s = SI13TeV. Gluinos, because of their strong color coupling, have the highest theoretical cross section of the sparticles that could be found at the LHC.

The decay topology of the simplified model for g̃→ tbarstopone → tt̄χ̃₁⁰ for both (a) on-shell and (b) off-shell stops, stopone. The difference between the two diagrams is that in the off-shell diagram, the stops are integrated out of the simplified model by setting the mass of the stop to SI5TeV.

The decay topology of the simplified model for g̃→ tbarstopone → tt̄χ̃₁⁰ for both (a) on-shell and (b) off-shell stops, stopone. The difference between the two diagrams is that in the off-shell diagram, the stops are integrated out of the simplified model by setting the mass of the stop to SI5TeV.

[102] A diagram of the LHC which sits on the border between Switzerland and France, near the city of Geneva. There are four main experiments located here: ALICE, ATLAS, CMS, and LHCb. All of the experimental apparatus are located around SI100meter underground where the stable rock is located. Also shown here is the Super Proton Synchotron which is the second-to-last stage of the injector chain before the particles enter the LHC tunnel.

A diagram of LHC injector complex [101,108]. Protons are accelerated in the following chain: Linac2→Proton Synchotron Booster→Proton Synchotron→Super Proton Synchotron→Large Hadron Collider.

LHC and High Luminosity LHC plan [109].

[110] Cumulative luminosity versus time delivered to (green) and recorded by ATLAS (yellow) during stable beams for proton-proton collisions at √s=SI13TeV center-of-mass energy in (a) 2015 and (b) 2016. The difference between delivered and recorded luminosity reflects the small inefficiency of the data acquisition in ATLAS. The uncertainty in the combined 2015+2016 integrated luminosity is 2.1%. It is derived, following a methodology similar to that detailed in [111], from a calibration of the luminosity scale using x-y beam-separation scans performed in August 2015 and May 2016.

[110] Cumulative luminosity versus time delivered to (green) and recorded by ATLAS (yellow) during stable beams for proton-proton collisions at √s=SI13TeV center-of-mass energy in (a) 2015 and (b) 2016. The difference between delivered and recorded luminosity reflects the small inefficiency of the data acquisition in ATLAS. The uncertainty in the combined 2015+2016 integrated luminosity is 2.1%. It is derived, following a methodology similar to that detailed in [111], from a calibration of the luminosity scale using x-y beam-separation scans performed in August 2015 and May 2016.

[110] The peak instantaneous luminosity delivered to ATLAS during stable beams for proton-proton collisions at √s=SI13TeV center-of-mass energy is shown for each LHC fill as a function of time in 2016. The luminosity is determined using counting rates measured by the luminosity detectors.

[110] The distribution of the mean number of interactions per crossing for the 2015-2016 proton-proton collision data at √s = SI13TeV center-of-mass energy. All data delivered to ATLAS during stable beams is shown, but not necessarily all of this data is suitable for an analysis.

[103] A cut-away view of the ATLAS detector. The dimensions of the detector are SI25meter in height and SI44meter in length. The overall weight of the detector is approximately SI7000tonne. This figure groups up major instrumentation components of the detector. Two people in red are shown for scale just to the right of the muon chambers on the left side of the figure.

[103] A slice of the ATLAS detector depicting the various particle interactions with each component of the detector. Dashed tracks in this figure are invisible to the detector component that the line is overlaid on top of. Muon track (orange) and neutrinos (dashed, white) pass through the entire detector. Electrons (yellow/green), photons (yellow/green), and hadrons (red/yellow) are fully absorbed by the calorimeter system. Charged particles like protons, electrons, and muons are curved by the solenoid magnet within the tracking system.

A top-down cartoon of the LHC, the SPS, and the four experiments at the LHC: ATLAS, CMS, LHCb, and ALICE. A common coordinate system is used in ATLAS. The positive x-axis points from the interaction point at the center of the ATLAS detector to the center of the LHC ring. the positive y-axis points from the interaction point upward to the surface of the earth. The z-axis runs along the beam line, with the detector half at positive z-values referred to as the ``A-side" (Geneva side) and the detector half at negative z-values referred to as the ``C-side" (Jura side).

A cartoon representation of selected pseudorapidity (η) values overlaid on cartesian geometry axes (dashed black lines). Red lines are drawn for η = ± 0.5, 1.0, 3.0.

[103] A longitudinal view of the ID compromising of the three main pieces: pixel detector, semiconductor tracker, and transition radiation tracker. It is SI6.2meter in length and SI2.1meter in height.

[103] Drawing showing the sensors and structural elements traversed by two charged tracks of p_T = SI10GeV in the end-cap ID at η = 1.4 and η = 2.2. The track at η=1.4 passes through the beam pipe, the three pixel layers, four SCT disks with double layers, and approximately 40 straws TRT end-cap. A particle at η=2.2 passes through the beam, only the first pixel layer, two of the end-cap pixel disks, and the last four end-cap SCT disks, but does not pass through the TRT straws which covers |η|leq2.0.

[103] A radial view of the ID with the detector elements crossed by a charged particle p_T = SI10GeV. The track pases through the beam pipe, three pixel layers, four SCT disks with double layers, and approximately TRT 40 straws.

[103] A longitudinal, cut-away view of the ATLAS calorimeter system showing: the tile calorimeter barrel, the tile calorimeter extended barrel, the liquid argon electromagnetic barrel, the liquid argon electromagnetic end-cap, the liquid argon hadronic end-cap, and the forward calorimeter.

[103] This is a sketch of a LAr accordion module where the different layers are visible in φ which is pointing up in this figure. The granularity in η and φ of the calorimeter cells for each of the three sampling layers and of the trigger towers (η × φ = 0.1 × 0.1) is also shown. These trigger towers will be discussed more in the crefchap:trigger section.

[103] This is a sketch of a Tile module showing how the mechanical assemply and the optical readout are integrated together. The various components of the optical readout are shown: the tiles, the fibers, and the photomultiplier tubes. Each wedge is approximately Δφ = 0.1 which is around SI20centimeter.

[103] A cut-away view of the ATLAS muon system.

[130] A schematic overview of the ATLAS TDAQ system in Run 2 with emphasis on the components relevant for triggering. The main subsystems shown are the L1 trigger, the HLT, and the FTK.

[130] A schematic view of the trigger towers used as input to the L1Calo trigger algorithms.

[179] The L1Calo system following the completion of the Phase-I upgrade at the start of Run 3. The new elements include the PPM, Optical Plant, Hub, ROD, and the three plFEX: eFEX, jFEX, and gFEX.

[179] A block diagram of the gFEX module. Shown are the real-time (to L1Topo) and readout (to FELIX) data paths. The Zynq+ is also shown.

A picture of the gFEX v4 board at Brookhaven National Lab. The three plpFPGA have not been placed yet. The Zynq+ is placed with a heat-sink and fan on top in the lower-right. This board has successfully loaded the custom Linux kernel I built along with the ``iroman" slow-control and monitoring software.

[179] gTower segmentation in the calorimeter. Note the special gTower coverage at 2.4 < |η| < 2.5 and 3.1 < |η| < 3.2. The central regions covered by pFPGA 1 and pFPGA 2 contain primarily plgTower of size up to Δη×Δφ = 0.2×0.2 and the forward regions have up to Δη×Δφ = 0.4×0.4. gBlock formation is also illustrated in colors here as contiguous blocks of plgTower, often 3 × 3. Note that plgBlock in an event are allowed to overlap. Also note that not all plgBlock have the same size.

gFEX rates shown for a simulated sample with (a) 80 and (b) 200 proton-proton interactions, representing potential data events for HL-LHC. Each curve corresponds to a different trigger item requiring one, two, or more gFEX trigger objects (plgTower [dark curves], plgBlock [light curves]) with the x-axis representing the energy threshold required for the trigger object and the y-axis showing the potential rate for the given selection. A lower threshold can provide a highly efficient trigger while maintaining rates based on the readout capabilities of the ATLAS detector.

gFEX rates shown for a simulated sample with (a) 80 and (b) 200 proton-proton interactions, representing potential data events for HL-LHC. Each curve corresponds to a different trigger item requiring one, two, or more gFEX trigger objects (plgTower [dark curves], plgBlock [light curves]) with the x-axis representing the energy threshold required for the trigger object and the y-axis showing the potential rate for the given selection. A lower threshold can provide a highly efficient trigger while maintaining rates based on the readout capabilities of the ATLAS detector.

Demonstration of ironman's flexibility and adaptability to various programming languages and data transfer models. The flexibility is demonstrated by (a) a website with Javascript polling the board, (b) direct access over SSH with Python, and (c) a RESTful server running on the board.

Demonstration of ironman's flexibility and adaptability to various programming languages and data transfer models. The flexibility is demonstrated by (a) a website with Javascript polling the board, (b) direct access over SSH with Python, and (c) a RESTful server running on the board.

Demonstration of ironman's flexibility and adaptability to various programming languages and data transfer models. The flexibility is demonstrated by (a) a website with Javascript polling the board, (b) direct access over SSH with Python, and (c) a RESTful server running on the board.

[181] The evolution of the partonic system which demonstrates how jets are formed and measured by the calorimeters in ATLAS.

[182] A candidate supersymmetry event with 6 jets (shown as cones) and a muon track which is colored red. The b-quark tagged jets are colored blue, while the non-b-quark tagged jets are colored yellow. The size of the jet cones are proportional to the measured and calibrated p_T of the jet. ID tracks with p_T > SI1GeV are colored green and their brightness is proportional to their p_T. The LAr and Tile are colored teal and yellow, respectively, and their length is proportional to the measured transverse energy deposit.

[186] Illustration of the (a) infrared sensitivity and (b) collinear sensitivty of a cursory designed jet algorithm.

[186] Illustration of the (a) infrared sensitivity and (b) collinear sensitivty of a cursory designed jet algorithm.

[189] A sample parton-level monte-carlo simulated event illustrating the shapes and areas of the resulting hard jets for R=1.0 and (a) P=1, (b) P=0, and (c) P=-1.

[189] A sample parton-level monte-carlo simulated event illustrating the shapes and areas of the resulting hard jets for R=1.0 and (a) P=1, (b) P=0, and (c) P=-1.

[189] A sample parton-level monte-carlo simulated event illustrating the shapes and areas of the resulting hard jets for R=1.0 and (a) P=1, (b) P=0, and (c) P=-1.

[193] The energy-equivalent cell noise in the ATLAS calorimeters on the electromagnetic scale as a function of |η| in the detector for the (a) 2010 configuration with ⟨ μ ⟩ = 0, (b) 2011 configuration with ⟨ μ ⟩ = 8, and (c) 2012 configuration with ⟨ μ ⟩ = 30. The various colors indicate the noise in the first layer of the FCal and the three layers of the LAr EMCal, the three layers of the Tile, the four layers of the HEC, and the three modules of the FCal. The higher levels in 2011 and 2012 compared to 2010 indicate the contribution from increased pile-up.

[193] The energy-equivalent cell noise in the ATLAS calorimeters on the electromagnetic scale as a function of |η| in the detector for the (a) 2010 configuration with ⟨ μ ⟩ = 0, (b) 2011 configuration with ⟨ μ ⟩ = 8, and (c) 2012 configuration with ⟨ μ ⟩ = 30. The various colors indicate the noise in the first layer of the FCal and the three layers of the LAr EMCal, the three layers of the Tile, the four layers of the HEC, and the three modules of the FCal. The higher levels in 2011 and 2012 compared to 2010 indicate the contribution from increased pile-up.

[193] The energy-equivalent cell noise in the ATLAS calorimeters on the electromagnetic scale as a function of |η| in the detector for the (a) 2010 configuration with ⟨ μ ⟩ = 0, (b) 2011 configuration with ⟨ μ ⟩ = 8, and (c) 2012 configuration with ⟨ μ ⟩ = 30. The various colors indicate the noise in the first layer of the FCal and the three layers of the LAr EMCal, the three layers of the Tile, the four layers of the HEC, and the three modules of the FCal. The higher levels in 2011 and 2012 compared to 2010 indicate the contribution from increased pile-up.

[193] Stages of topocluster formation in the first module of the FCal for a simulated dijet event with at least one jet entering the calorimeter is shown for √s = SI8TeV. Shown are (a) seed cells for topoclustering, (b) growth cells, and (c) boundary cells. Cells which are not colored, but contained inside a topocluster boundary are cells with negative energy.

[193] Stages of topocluster formation in the first module of the FCal for a simulated dijet event with at least one jet entering the calorimeter is shown for √s = SI8TeV. Shown are (a) seed cells for topoclustering, (b) growth cells, and (c) boundary cells. Cells which are not colored, but contained inside a topocluster boundary are cells with negative energy.

[193] Stages of topocluster formation in the first module of the FCal for a simulated dijet event with at least one jet entering the calorimeter is shown for √s = SI8TeV. Shown are (a) seed cells for topoclustering, (b) growth cells, and (c) boundary cells. Cells which are not colored, but contained inside a topocluster boundary are cells with negative energy.

[196] Calibration stages for EM-scale jets. Other than the origin correction, each stage of the calibration is applied to the four-momentum of the jet.

[196] Dependence of the EM-scale anti-k_t jet p_T on (a) in-time pile-up and (b) out-of-time pile-up as a function of |η| for p_T^truth = SI25GeV. The dependence is shown before pile-up corrections (blue), after area-based pile-up correction (purple), and after residual corrections (red) using N_PV and μ.

[196] Dependence of the EM-scale anti-k_t jet p_T on (a) in-time pile-up and (b) out-of-time pile-up as a function of |η| for p_T^truth = SI25GeV. The dependence is shown before pile-up corrections (blue), after area-based pile-up correction (purple), and after residual corrections (red) using N_PV and μ.

[196] The average jet energy response as a function of η_det for jets of a truth energy of 30, 60, 110, 400, and 1200 GeV. The energy response is shown after origin and pile-up corrections are applied. This shows the size of the calibration constants for jets built from topoclusters at the EM scale.

[196] The signed difference between truth jet η^truth and the reconstructed jet η^reco due to biases in the jet reconstruction. This bias is addressed with an η correction applied as a function of |η_det|. The effect of changing geometries, such as barrel end-cap transition region around |η_det ∼ 1.4| and endcap-forward transition region around |η_det| ∼ 3.1 can clearly be seen.

[196] The average jet response in Monte-Carlo simulation as a function of the GSC variables for three different ranges of p_T^truth. (a) and (b) are shown with no GSC corrections applied. (c) and (d) are shown with the first two corrections applied. (e) is shown with the previous four corrections applied. (a)-(d) are shown for |η_det| < 0.1 while (e) is shown for |η_det| < 1.3.

[196] The average jet response in Monte-Carlo simulation as a function of the GSC variables for three different ranges of p_T^truth. (a) and (b) are shown with no GSC corrections applied. (c) and (d) are shown with the first two corrections applied. (e) is shown with the previous four corrections applied. (a)-(d) are shown for |η_det| < 0.1 while (e) is shown for |η_det| < 1.3.

[196] The average jet response in Monte-Carlo simulation as a function of the GSC variables for three different ranges of p_T^truth. (a) and (b) are shown with no GSC corrections applied. (c) and (d) are shown with the first two corrections applied. (e) is shown with the previous four corrections applied. (a)-(d) are shown for |η_det| < 0.1 while (e) is shown for |η_det| < 1.3.

[196] The average jet response in Monte-Carlo simulation as a function of the GSC variables for three different ranges of p_T^truth. (a) and (b) are shown with no GSC corrections applied. (c) and (d) are shown with the first two corrections applied. (e) is shown with the previous four corrections applied. (a)-(d) are shown for |η_det| < 0.1 while (e) is shown for |η_det| < 1.3.

[196] The average jet response in Monte-Carlo simulation as a function of the GSC variables for three different ranges of p_T^truth. (a) and (b) are shown with no GSC corrections applied. (c) and (d) are shown with the first two corrections applied. (e) is shown with the previous four corrections applied. (a)-(d) are shown for |η_det| < 0.1 while (e) is shown for |η_det| < 1.3.

[196] The average jet p_T response of EM+JES jets calibrated up to the η-intercalibration as a function of jet p_T for (a) Z+jet events, (b) γ+jet events, and (c) multijet balance.

[196] The average jet p_T response of EM+JES jets calibrated up to the η-intercalibration as a function of jet p_T for (a) Z+jet events, (b) γ+jet events, and (c) multijet balance.

[196] The average jet p_T response of EM+JES jets calibrated up to the η-intercalibration as a function of jet p_T for (a) Z+jet events, (b) γ+jet events, and (c) multijet balance.

[196] The in-situ calibrations for Z+jet, γ+jet, and multijet balance are statistically combined to create a total calibration and uncertainty. The final derived correction (black line) and its statistical uncertainty band (dark blue) and total uncertainty band (light green) are shown.

[196] Combined uncertainty of JES calibration as a function of (a) jet p_T at η = 0 and (b) η at p_T = SI80GeV.

[196] Combined uncertainty of JES calibration as a function of (a) jet p_T at η = 0 and (b) η at p_T = SI80GeV.

[199] Jet kinematic distributions for η_det for jets above p_T > SI25GeV. The Monte-Carlo prediction is normalized to the data and the error s reflect the statistical uncertainty.

[199] Topocluster multiplicity for jets with p_T > SI25GeV is shown as a function of jet η. The error bars reflect the statistical uncertainty.

[199] The minimum distance in ΔR_cone to a jet with p_T > SI25GeV and |y| < 0.8. The Monte-Carlo prediction is normalized to the data. Requiring R=0.4 jets to be isolated requires the minimum ΔR_cone > 1.0 = 2.5 × R.

[205] The (a) transverse and (b) longitudinal signed impact parameter significance of tracks in tt̄ events for b-flavor (blue), c-flavor (green), and light flavour (red) jets. The tail shown in the b jets line (blue) is due to the long lifetime of B-hadrons.

[205] The (a) transverse and (b) longitudinal signed impact parameter significance of tracks in tt̄ events for b-flavor (blue), c-flavor (green), and light flavour (red) jets. The tail shown in the b jets line (blue) is due to the long lifetime of B-hadrons.

[205] The log-likelihood ratio for the (a) IP2D and (b) IP3D b-tagging algorithms in tt̄ events for b-flavor (blue), c-flavor (green), and light flavour (red) jets. The log-likelihood is calculated as the ratio of b-flavor and light-flavor probability density functions. Jets with no tracks are assigned a large negative value in an underflow bin which is not shown on this plot.

[205] The log-likelihood ratio for the (a) IP2D and (b) IP3D b-tagging algorithms in tt̄ events for b-flavor (blue), c-flavor (green), and light flavour (red) jets. The log-likelihood is calculated as the ratio of b-flavor and light-flavor probability density functions. Jets with no tracks are assigned a large negative value in an underflow bin which is not shown on this plot.

[205] Properties of the secondary vertices reconstructed by the SV algorithm for b-flavor (blue), c-flavor (green), and light flavour (red) jets. Shown are (a) the invariant mass of the tracks associated with the reconstructed vertex and (b) the number of two-track vertices reconstructed within the jet.

[205] Properties of the secondary vertices reconstructed by the SV algorithm for b-flavor (blue), c-flavor (green), and light flavour (red) jets. Shown are (a) the invariant mass of the tracks associated with the reconstructed vertex and (b) the number of two-track vertices reconstructed within the jet.

[205] Properties of the secondary vertices reconstructed by the JF algorithm for b-flavor (blue), c-flavor (green), and light flavour (red) jets. Shown are (a) the invariant mass of the tracks associated with the reconstructed vertex and (b) the number of two-track vertices reconstructed within the jet.

[205] Properties of the secondary vertices reconstructed by the JF algorithm for b-flavor (blue), c-flavor (green), and light flavour (red) jets. Shown are (a) the invariant mass of the tracks associated with the reconstructed vertex and (b) the number of two-track vertices reconstructed within the jet.

[205] (a) Light-flavor jet and (b) c-flavor jet rejection versus b-flavor jet efficiency for the 2015 and 2016 configurations of the MV2 b-tagging algorithm is shown evaluated for tt̄ events.

[205] (a) Light-flavor jet and (b) c-flavor jet rejection versus b-flavor jet efficiency for the 2015 and 2016 configurations of the MV2 b-tagging algorithm is shown evaluated for tt̄ events.

[205] The MV2c10 output for b-flavor (blue), c-flavor (green), and light flavour (red) jets evaluated with tt̄ events.

[213] The data/simulation scale factors for the MV1 algorithm at 70% b-jet tagging efficiency. The error bars show the statistical uncertainties while the green band indicates the total uncertainty. Please note that the algorithm here is different from the MV2 algorithm used in this thesis analysis. The performance studies for the current algorithm are still being done.

[214] Total uncertainty in the efficiency scale factor for ``medium" muons as a function of p_T as obtained from Z→μμ (solid lines) and J/Ψ → μμ (dashed lines) decays. The combined uncertainty is the sum in quadrature of the individual contributions.

[222] The distribution of reconstructed track soft-term E_T^miss is shown for (a) Z → μμ topologies and (b) W → e ν_e topologies. The agreement between data and simulation for W → e ν_e is notably worse with respect to Z → μμ in the low E_T^miss region likely due to the missing QCD multijet background not included in the studies.

[222] The distribution of reconstructed track soft-term E_T^miss is shown for (a) Z → μμ topologies and (b) W → e ν_e topologies. The agreement between data and simulation for W → e ν_e is notably worse with respect to Z → μμ in the low E_T^miss region likely due to the missing QCD multijet background not included in the studies.

A feynman diagram of the two-body decay of a neutral pion π⁰ to two photons γ in the (a) lab frame and (b) rest frame.

A feynman diagram of the two-body decay of a neutral pion π⁰ to two photons γ in the (a) lab frame and (b) rest frame.

[223] The angular separation between the (a) W boson and b-quark in top decays and (b) light quarks in W boson decays as a function of the p_T of the parent particle. Both distributions are at the generator level and do not include effects due to initial and final-state radiation, or the underlying event.

[223] The angular separation between the (a) W boson and b-quark in top decays and (b) light quarks in W boson decays as a function of the p_T of the parent particle. Both distributions are at the generator level and do not include effects due to initial and final-state radiation, or the underlying event.

The decay topology in the off-shell Gtt simplified model.

Distributions of the Δφ_min^4j variable before and after jet smearing is applied for 2015-2016 data using a 0-lepton preselection, without the Δφ_min^4j selection applied. The top panel shows the distribution in log-scale of the number of events in each bin of Δφ_min^4j. The second panel shows the background composition with each background contribution some fraction of the total background. The third panel shows the data/MC ratio where data is compared to the total estimated, pre-fit background. Looking at the background composition, it is clear that the multi-jet background (red) is reducible, as is evidenced by a Δφ_min^4j > 0.4 selection that can be applied.

Distributions of the Δφ_min^4j variable before and after jet smearing is applied for 2015-2016 data using a 0-lepton preselection, without the Δφ_min^4j selection applied. The top panel shows the distribution in log-scale of the number of events in each bin of Δφ_min^4j. The second panel shows the background composition with each background contribution some fraction of the total background. The third panel shows the data/MC ratio where data is compared to the total estimated, pre-fit background. Looking at the background composition, it is clear that the multi-jet background (red) is reducible, as is evidenced by a Δφ_min^4j > 0.4 selection that can be applied.

[239] Efficiency of the 2015 E_T^miss trigger for different requirements on the leading jet p_T. The muons are subtracted from E_T^miss to reproduce the L1 E_T^miss. The reference trigger used to select events is HLT_mu26_medium.

[239] Efficiency of the 2015 E_T^miss trigger for different requirements on the leading jet p_T. The muons are subtracted from E_T^miss to reproduce the L1 E_T^miss. The reference trigger used to select events is HLT_mu26_medium.

[239] Efficiency of the various 2016 E_T^miss triggers as a function of the offline E_T^miss. The muons are subtracted from E_T^miss to reproduce the L1 E_T^miss. The reference trigger used to select events is HLT_mu26_ivarmedium. One of the triggers used in the analysis for early 2016 data is shown here: HLT_xe100_mht_L1XE50.

[239] Efficiency of the various 2016 E_T^miss triggers as a function of the offline E_T^miss. The muons are subtracted from E_T^miss to reproduce the L1 E_T^miss. The reference trigger used to select events is HLT_mu26_ivarmedium. One of the triggers used in the analysis for early 2016 data is shown here: HLT_xe100_mht_L1XE50.

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.endfigure

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.endfigure

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.endfigure

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.endfigure

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.

Distributions of discriminating variables for events passing the 0-lepton preselection criteria. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. endfigure

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. endfigure

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. endfigure

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison. endfigure

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.

Distributions of discriminating variables for events passing the 1-lepton preselection criteria, after applying the kinematic reweighting to the m_eff distribution described in the text. The statistical and experimental systematic uncertainties (as defined in crefsec:systematics) are included in the uncertainty band. The last bin includes overflow events. The lower part of each figure shows the ratio of data to the background prediction. All backgrounds (including tt̄) are normalized using the best available theoretical calculation described in crefsec:trigger:samples. The background category tt̄+X includes tt̄ W/Z, tt̄ h and tt̄ tt̄ events. Example signal models with cross-sections enhanced by a factor of 50 are overlaid for comparison.

A diagramatic overview of how the different regions SR, CR, and VR are defined for the cut-and-count analysis

Significance of optimal cuts (creffig:optGtt0L_fullGrid_cuts) for each grid point in the Gtt 0-lepton channel. The grid represents all the mass points for the g̃-χ̃₁⁰ pair with the mass of the gluino on the x-axis in GeV and the mass of the neutralino on the y-axis in GeV. This optimization was performed at an assumed total integrated luminosity of SI35fb^-1. The Run-2 limit from my first paper on this search at SI3.6fb^-1 is overlaid in red [230]. The significance reported is an estimate of the CLs method using BinomialExpZ of RooStats [241,242]. This significance represents the discriminating power the optimal cut has in each mass point.

Optimal cut values for each grid point in the Gtt 0-lepton channel. The grid represents all the mass points for the g̃-χ̃₁⁰ pair with the mass of the gluino on the x-axis in GeV and the mass of the neutralino on the y-axis in GeV. This optimization was performed at an assumed total integrated luminosity of SI35fb^-1. The kinematic variables were scanned over the values defined in creftab:optimizationscan0L, a significance for each combination of selections was computed, and the selection that provided maximum discriminating power in each mass point is plotted. Each plot represents the cut applied to the kinematic variable (a) m_eff, (b) M_J^Σ, (c) E_T^miss, (d) N_b-jets.

Optimal cut values for each grid point in the Gtt 0-lepton channel. The grid represents all the mass points for the g̃-χ̃₁⁰ pair with the mass of the gluino on the x-axis in GeV and the mass of the neutralino on the y-axis in GeV. This optimization was performed at an assumed total integrated luminosity of SI35fb^-1. The kinematic variables were scanned over the values defined in creftab:optimizationscan0L, a significance for each combination of selections was computed, and the selection that provided maximum discriminating power in each mass point is plotted. Each plot represents the cut applied to the kinematic variable (a) m_eff, (b) M_J^Σ, (c) E_T^miss, (d) N_b-jets.

Optimal cut values for each grid point in the Gtt 0-lepton channel. The grid represents all the mass points for the g̃-χ̃₁⁰ pair with the mass of the gluino on the x-axis in GeV and the mass of the neutralino on the y-axis in GeV. This optimization was performed at an assumed total integrated luminosity of SI35fb^-1. The kinematic variables were scanned over the values defined in creftab:optimizationscan0L, a significance for each combination of selections was computed, and the selection that provided maximum discriminating power in each mass point is plotted. Each plot represents the cut applied to the kinematic variable (a) m_eff, (b) M_J^Σ, (c) E_T^miss, (d) N_b-jets.

Optimal cut values for each grid point in the Gtt 0-lepton channel. The grid represents all the mass points for the g̃-χ̃₁⁰ pair with the mass of the gluino on the x-axis in GeV and the mass of the neutralino on the y-axis in GeV. This optimization was performed at an assumed total integrated luminosity of SI35fb^-1. The kinematic variables were scanned over the values defined in creftab:optimizationscan0L, a significance for each combination of selections was computed, and the selection that provided maximum discriminating power in each mass point is plotted. Each plot represents the cut applied to the kinematic variable (a) m_eff, (b) M_J^Σ, (c) E_T^miss, (d) N_b-jets.

((a)) Optimal SR and ((b)) significance for the optimal region for all points of the grid in the Gtt 0-lepton analysis.

((a)) Optimal SR and ((b)) significance for the optimal region for all points of the grid in the Gtt 0-lepton analysis.

Significance of each SR at each grid point of the Gtt 0-lepton analysis.

Significance of each SR at each grid point of the Gtt 0-lepton analysis.

Significance of each SR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

Signal contamination of each VR at each grid point of the Gtt 0-lepton analysis.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L boosted regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L boosted regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L boosted regions.

N-1 plot of the total jet mass variable in the 0-lepton boosted region. The background is stacked in the histogram and three signal points are overlaid to show the shape comparisons, all normalized to SI35fb^-1 except the highest mass signal curve which has been scaled up by a factor of 10.

Significance of optimal cuts for each grid point in the Gtt 0-lepton channel.

Optimal cut values for each grid point in the Gtt 1-lepton channel.

Optimal cut values for each grid point in the Gtt 1-lepton channel.

Optimal cut values for each grid point in the Gtt 1-lepton channel.

Optimal cut values for each grid point in the Gtt 1-lepton channel.

((a)) Optimal SR and ((b)) significance for the optimal region for all points of the grid in the Gtt 1-lepton analysis.

((a)) Optimal SR and ((b)) significance for the optimal region for all points of the grid in the Gtt 1-lepton analysis.

Significance of each SR at each grid point of the Gtt 1-lepton analysis.

Significance of each SR at each grid point of the Gtt 1-lepton analysis.

Significance of each SR at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each CR at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_T at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_T at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_T at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_{T, min}^b-jets at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_{T, min}^b-jets at each grid point of the Gtt 1-lepton analysis.

Signal contamination of each VR-m_{T, min}^b-jets at each grid point of the Gtt 1-lepton analysis.

Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

N-1 plot of the total jet mass variable in the 1-lepton boosted region. The background is stacked in the histogram and three signal points are overlaid to show the shape comparisons, all normalized to SI35fb^-1, except for the highest mass signal curve which has been scaled up by a factor of 10.

The value of μ_tt̄, and uncertainty, after the background-only fit, for the regions defined in crefsec:cutandcount.

Relative systematic uncertainty in the background estimate for the cut-and-count analysis. The individual uncertainties can be correlated, such that the total background uncertainty is not necessarily their sum in quadrature.

The summary of the calculated theory uncertainties by comparison of generator predictions is shown here. Two different versions are shown, without and with truth b-tagging. One of the main problems was obtaining enough statistical power in truth generator samples. Instead of vetoing truth events that do not have enough b-tags, one uses truth b-tagging which weights each event by the probability for it to have the given N_b-jets requirement (both inclusively and exclusively). Each component of the uncertainty is shown as stacked and the sum in quadrature is overlaid in black. The uncertainty of the systematic is shaded red. The unit is the full size of the uncertainty (i.e. 2 corresponds to a 200% uncertainty).

The summary of the calculated theory uncertainties by comparison of generator predictions is shown here. Two different versions are shown, without and with truth b-tagging. One of the main problems was obtaining enough statistical power in truth generator samples. Instead of vetoing truth events that do not have enough b-tags, one uses truth b-tagging which weights each event by the probability for it to have the given N_b-jets requirement (both inclusively and exclusively). Each component of the uncertainty is shown as stacked and the sum in quadrature is overlaid in black. The uncertainty of the systematic is shaded red. The unit is the full size of the uncertainty (i.e. 2 corresponds to a 200% uncertainty).

The summary of the calculated theory uncertainties for single top is shown here. Two different versions are shown, without and with truth b-tagging. One of the main problems was obtaining enough statistical power in truth generator samples. Instead of vetoing truth events that do not have enough b-tags, one uses truth b-tagging which weights each event by the probability for it to have the given N_b-jets requirement (both inclusively and exclusively). Each component of the uncertainty is shown as stacked and the sum in quadrature is overlaid in black. The uncertainty of the systematic is shaded red. The unit is the full size of the uncertainty (i.e. 2 corresponds to a 200% uncertainty).

The summary of the calculated theory uncertainties for single top is shown here. Two different versions are shown, without and with truth b-tagging. One of the main problems was obtaining enough statistical power in truth generator samples. Instead of vetoing truth events that do not have enough b-tags, one uses truth b-tagging which weights each event by the probability for it to have the given N_b-jets requirement (both inclusively and exclusively). Each component of the uncertainty is shown as stacked and the sum in quadrature is overlaid in black. The uncertainty of the systematic is shaded red. The unit is the full size of the uncertainty (i.e. 2 corresponds to a 200% uncertainty).

Results of the background-only fit extrapolated to the validation regions of the cut-and-count analysis. The tt̄ normalization, μ_tt̄, is obtained from the fit to the control regions shown in creffig:mu_ttbar. The upper panel shows the observed number of events and the predicted background yield. All uncertainties are included in the uncertainty band. The background category tt̄+X includes tt̄+W/Z, tt̄+H and tt̄tt̄ events. The lower panel shows the pulls in each validation region.

Results of the background-only fit extrapolated to the unblinded signal regions of the cut-and-count analysis. The tt̄ normalization, μ_tt̄, is obtained from the fit to the control regions shown in creffig:mu_ttbar. The data in the signal regions are not included in the fit. The upper panel shows the observed number of events and the predicted background yield. All uncertainties are included in the uncertainty band. The background category tt̄+X includes tt̄+W/Z, tt̄+H and tt̄tt̄ events. The lower panel shows the pulls in each signal region.

Example Poissonian probability density functions for background (blue) and signal+background (red) hypotheses for 10 observed events (black line). In this example, p(10 | b) = 0.018 and p(10 | s+b) = 0.125. The observed data is more likely under the s+b hypothesis than background-only.

Exclusion limits in the χ̃₁⁰ and g̃ mass plane for the Gtt model obtained in the context of the cut-and-count analysis. The dashed and solid bold lines show the 95% CL expected and observed limits, respectively. The shaded bands around the expected limits show the impact of the experimental and background uncertainties. The dotted lines show the impact on the observed limit of the variation of the nominal signal cross-section by ± 1σ of its theoretical uncertainty. The 95% CL expected and observed limits from the ATLAS search based on 2015 data [252] are also shown.

[253,254] Exclusion limits at 95% CL based on SI13TeV data for (a) ATLAS and (b) CMS in the (g̃, χ̃₁⁰) mass plane for different simplified models featuring the decay of the gluino to the lightest supersymmetric particle (lightest neutralino or gravitino) either directly or through a cascade chain featuring other SUSY particles with intermediate measses. For each line, the gluino decay mode is reported in the legend, along with the arXiv reference, and it is assumed to proceed with 100% branching ratio. Some limits depend on additional assumptions on the mass of the intermediate states, as described in the references provided in the plot (ATLAS [255,182,256,257,258,259,260]; CMS [261,262,263,264,265,266]). The search presented in this thesis is shown in magenta for ATLAS.

[253,254] Exclusion limits at 95% CL based on SI13TeV data for (a) ATLAS and (b) CMS in the (g̃, χ̃₁⁰) mass plane for different simplified models featuring the decay of the gluino to the lightest supersymmetric particle (lightest neutralino or gravitino) either directly or through a cascade chain featuring other SUSY particles with intermediate measses. For each line, the gluino decay mode is reported in the legend, along with the arXiv reference, and it is assumed to proceed with 100% branching ratio. Some limits depend on additional assumptions on the mass of the intermediate states, as described in the references provided in the plot (ATLAS [255,182,256,257,258,259,260]; CMS [261,262,263,264,265,266]). The search presented in this thesis is shown in magenta for ATLAS.

For the Gtt 0-lepton boosted region, (a) signal acceptance at truth level, (b) signal acceptance otimes efficiency at the reconstruction level, and (c) calculated signal efficiency are shown in the (g̃, χ̃₁⁰) mass plane. The z-axis represents the value of each bin in units % with 0% being yellow, and 100% (25% for acceptance) being green.

For the Gtt 0-lepton boosted region, (a) signal acceptance at truth level, (b) signal acceptance otimes efficiency at the reconstruction level, and (c) calculated signal efficiency are shown in the (g̃, χ̃₁⁰) mass plane. The z-axis represents the value of each bin in units % with 0% being yellow, and 100% (25% for acceptance) being green.

For the Gtt 0-lepton boosted region, (a) signal acceptance at truth level, (b) signal acceptance otimes efficiency at the reconstruction level, and (c) calculated signal efficiency are shown in the (g̃, χ̃₁⁰) mass plane. The z-axis represents the value of each bin in units % with 0% being yellow, and 100% (25% for acceptance) being green.

[268] Illustration of a simulated large radius anti-k_t jet, with R=1.0 from a top quark produced in a Z' → tt̄ decay with m_Z' = SI1.75TeV. The (a) event display and (b) parton shower history for an example decay. Subjets are identified by a particular color in the event display: W boson (red), b-jet (green), top radiation (yellow), and initial state radiation (blue). Shown is a black circle representing the size of the R=1.0 anti-k_t offline jet that is clustered and can capture the full information of the top decay and a dashed, purple rectangular window of size 0.8times0.8 representing the L1 trigger sliding window algorithm for identifying energy above a fixed threshold.

[268] Illustration of a simulated large radius anti-k_t jet, with R=1.0 from a top quark produced in a Z' → tt̄ decay with m_Z' = SI1.75TeV. The (a) event display and (b) parton shower history for an example decay. Subjets are identified by a particular color in the event display: W boson (red), b-jet (green), top radiation (yellow), and initial state radiation (blue). Shown is a black circle representing the size of the R=1.0 anti-k_t offline jet that is clustered and can capture the full information of the top decay and a dashed, purple rectangular window of size 0.8times0.8 representing the L1 trigger sliding window algorithm for identifying energy above a fixed threshold.

[269] Jet mass for leading p_T anti-k_t trimmed jets with R=1.0, |η| < 1.2, and p_T > SI350GeV. Here, ``contained" refers to events having a hadronically-decaying top quark t with collimated daughter particles at the truth level (all three daughter quarks q_i ≡ bqqbar satisfy ΔR_cone(q_i, t) < 1.0). The shaded band represents the bin-by-bin statistical uncertainty in monte-carlo simulation.

A distribution of the angular variable ΔR_cone betwen the leading jet in the event and the leading gTower in the event for monte-carlo simulated tt̄ samples with ⟨ μ ⟩ = 80 at a center-of-mass energy of √s = SI14TeV. A majority of towers are found within ΔR_cone < 1.0 of the reconstructed, isolated offline jet in the event.

A canonical example that demonstrates the algorithms in the preceding subsections.

A canonical example that demonstrates the algorithms in the preceding subsections.

A canonical example that demonstrates the algorithms in the preceding subsections.

No glsplgTower were found for the given threshold of SI20GeV.

No glsplgTower were found for the given threshold of SI20GeV.

No glsplgTower were found for the given threshold of SI20GeV.

Overlapping glsplgTower with E_T > SI20GeV showing that even a crude version of a clustering algorithm is still able to identify at least two jets at almost identical locations.

Overlapping glsplgTower with E_T > SI20GeV showing that even a crude version of a clustering algorithm is still able to identify at least two jets at almost identical locations.

Overlapping glsplgTower with E_T > SI20GeV showing that even a crude version of a clustering algorithm is still able to identify at least two jets at almost identical locations.

Example distributions of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T (a) without a trigger selection applied and (b-d) a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the number of events. Turn-on curves from dividing each trigger-selected distribution by the denominator (no selection) is shown in creffig:upgrade:trigger:dennum:turnon.

Example distributions of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T (a) without a trigger selection applied and (b-d) a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the number of events. Turn-on curves from dividing each trigger-selected distribution by the denominator (no selection) is shown in creffig:upgrade:trigger:dennum:turnon.

Example distributions of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T (a) without a trigger selection applied and (b-d) a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the number of events. Turn-on curves from dividing each trigger-selected distribution by the denominator (no selection) is shown in creffig:upgrade:trigger:dennum:turnon.

Example distributions of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T (a) without a trigger selection applied and (b-d) a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the number of events. Turn-on curves from dividing each trigger-selected distribution by the denominator (no selection) is shown in creffig:upgrade:trigger:dennum:turnon.

Example distributions of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T (a) without a trigger selection applied and (b-d) a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the number of events. Turn-on curves from dividing each trigger-selected distribution by the denominator (no selection) is shown in creffig:upgrade:trigger:dennum:turnon.

Example turn-on curves of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T with a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the efficiency of the trigger. These curves were calculated from distributions in creffig:upgrade:trigger:dennum.

Example turn-on curves of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T with a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the efficiency of the trigger. These curves were calculated from distributions in creffig:upgrade:trigger:dennum.

Example turn-on curves of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T with a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the efficiency of the trigger. These curves were calculated from distributions in creffig:upgrade:trigger:dennum.

Example turn-on curves of reconstructed, uncalbrated, leading, anti-k_t R=1.0 offline jet p_T with a requirement on the leading gTower E_T. Offline jets are matched to the leading gTower in an event, so this amounts to an event-level trigger selection. The y-axis is the efficiency of the trigger. These curves were calculated from distributions in creffig:upgrade:trigger:dennum.

Example turn-on curves of different triggers to understand the impact of changes in the gFEX jet reconstruction algorithms for monte-carlo simulated tt̄ events with center-of-mass energy √s = SI14TeV, requiring the leading trigger jet to have p_T > SI140GeV. The gFEX jet reconstruction algorithm is seeded by a SI15GeV gTower to form gFEX trigger jets. The green curve represents the efficiency of this standard reconstruction. The red curve represents the efficiency of the trigger jets, after they have been corrected for estimated pile-up energy density in the event, described more in crefsec:upgrade:studies:pileup. Because there was an observed shift in the location of the trigger curve, the blue curve represents the efficiency using uncorrected trigger jets, but tightens the trigger selection from SI140GeV to SI240GeV (a shift of SI98GeV), to closely match the location of the red curve and understand the impact on the resolution of the trigger. Each turn-on curve is parameterized by the resolution w and the plateau x_0.95: green is (w, x_0.95) = (31.1, 179.3), blue is (w, x_0.95) = (40.7, 299.4), and red is (w, x_0.95) = (53.9, 316.8).

footnotesize[270] Per-jet efficiency turn-on curves in Monte Carlo (MC) simulation for multiple Phase I upgrade Level-1 jet trigger options. A global feature extraction (gFEX) reconstruction algorithm (closed red markers, left) from the TDAQ Phase I Upgrade Technical Design Report (TDR) [177] with a SI140GeV threshold is compared to full simulation of the Run I Level-1 calorimeter jet trigger (open blue markers, left and right) with a SI100GeV threshold. The gFEX reconstruction implements a simple seeded cone algorithm with a nominal radius of R=1.0 and with a seed selection of SI15GeV applied to calorimeter towers with area 0.2times0.2 in ηtimesφ. The SI140GeV gFEX trigger threshold is chosen to match the L1_J100 single subjet turn-on curve. Pair-produced top quark monte-carlo simulation samples are simulated with a pile-up level equivalent to an average number of interactions per bunch-crossing ⟨ μ ⟩ = 80. For each algorithm, the efficiency curves are shown as a function of the offline trimmed anti-k_t R=1.0 jet p_T with different offline subjet multiplicities. The trimming parameters specify that any subjets with a p_T fraction of the original jet less than 5% are to be discarded. The subjets are defined using the kt-clustering algorithm with a nominal radius parameter of D=0.3. For subjet counting, the subjets are required to have a subjet p_T>SI20GeV. The offline trimmed jets are required to be isolated from any other offline jet by at least a radial distance of ΔR_cone > SI2.0radian and to be within the pseudorapidity range |η| < 2.5. The turn-on curves measure per-jet efficiencies after requiring a that the the Level-1 gFEX jet be within ΔR_cone < 1.0 of the offline trimmed jet.

Distributions of truncated-mean-based online pile-up calculation using glsplgTower across different η ranges compared to offline pile-up (blue) for monte-carlo simulated tt̄ events with ⟨ μ ⟩ = 80 at center-of-mass energy √s = SI14TeV. While the scale is not important as this can be calibrated later, the width of each distribution is and how well it corresponds to offline. The upper threshold, X, also labeled on the plot as ρ(E_T^tower < X siGeV, is also specified as (a) SI3GeV and (b) SI6GeV.

Distributions of truncated-mean-based online pile-up calculation using glsplgTower across different η ranges compared to offline pile-up (blue) for monte-carlo simulated tt̄ events with ⟨ μ ⟩ = 80 at center-of-mass energy √s = SI14TeV. While the scale is not important as this can be calibrated later, the width of each distribution is and how well it corresponds to offline. The upper threshold, X, also labeled on the plot as ρ(E_T^tower < X siGeV, is also specified as (a) SI3GeV and (b) SI6GeV.

[270] Correlation between the offline event energy density ρ [271] on the horizontal axis and a simplified calculation of the event energy density in the L1Calo trigger using gFEX with a truncated-mean-based approach using plgTower with E_T^tower < SI6GeV and -1.6 < η < 0.0. The correlation for (a) tt̄ and (b) ZH → νν bbbar events is greater than 90%. Both monte-carlo simulation samples are simulated with average number of interactions ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. In each case, the strong correlation means that the average value of ρ measured by the gFEX trigger for a given offline ρ is similar.

[270] Correlation between the offline event energy density ρ [271] on the horizontal axis and a simplified calculation of the event energy density in the L1Calo trigger using gFEX with a truncated-mean-based approach using plgTower with E_T^tower < SI6GeV and -1.6 < η < 0.0. The correlation for (a) tt̄ and (b) ZH → νν bbbar events is greater than 90%. Both monte-carlo simulation samples are simulated with average number of interactions ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. In each case, the strong correlation means that the average value of ρ measured by the gFEX trigger for a given offline ρ is similar.

Correlation between isolated offline jet and matched gFEX trigger jet energies for -1.6 < η < 0.0 in a monte-carlo simulated tt̄ sample with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The correlations are shown (a) before and (b) after pile-up mitigation is applied. The trigger jets were seeded using towers with E_T > SI15GeV and the truncated-mean-based pile-up ρ_online was calculated using towers with E_T < SI6GeV which was optimized. The white circles represent the average trigger jet energy in each offline jet bin.

Correlation between isolated offline jet and matched gFEX trigger jet energies for -1.6 < η < 0.0 in a monte-carlo simulated tt̄ sample with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The correlations are shown (a) before and (b) after pile-up mitigation is applied. The trigger jets were seeded using towers with E_T > SI15GeV and the truncated-mean-based pile-up ρ_online was calculated using towers with E_T < SI6GeV which was optimized. The white circles represent the average trigger jet energy in each offline jet bin.

Correlation between isolated offline jet and the energy resolution of the matched gFEX trigger jet is shown for -1.6 < η < 0.0 in a monte-carlo simulated tt̄ sample with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The resolutions are shown (a) before and (b) after pile-up mitigation is applied. The trigger jets were seeded using towers with E_T > SI15GeV and the truncated-mean-based pile-up ρ_online was calculated using towers with E_T < SI6GeV which was optimized. The white circles represent the average resolution in each offline jet bin. The resolution of the trigger jet energy is defined as a measure of the difference with respect to the matched offline jet compared to the energy of the offline jet.

Correlation between isolated offline jet and the energy resolution of the matched gFEX trigger jet is shown for -1.6 < η < 0.0 in a monte-carlo simulated tt̄ sample with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The resolutions are shown (a) before and (b) after pile-up mitigation is applied. The trigger jets were seeded using towers with E_T > SI15GeV and the truncated-mean-based pile-up ρ_online was calculated using towers with E_T < SI6GeV which was optimized. The white circles represent the average resolution in each offline jet bin. The resolution of the trigger jet energy is defined as a measure of the difference with respect to the matched offline jet compared to the energy of the offline jet.

Y-axis Projections of the resolution plots in creffig:upgrade:studies:mitigation:resolution for selected offline jet p_T ranges: SIrange170180GeV, SIrange200220GeV, and SIrange300350GeV. This was done on monte-carlo simulated tt̄ events with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The projections are shown (a) before and (b) after pile-up mitigation is applied. Each legend also reports the full-width half-max (FWHM) of a Gaussian fit to each of the projections, a smaller value being a stronger resolution.

Y-axis Projections of the resolution plots in creffig:upgrade:studies:mitigation:resolution for selected offline jet p_T ranges: SIrange170180GeV, SIrange200220GeV, and SIrange300350GeV. This was done on monte-carlo simulated tt̄ events with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. The projections are shown (a) before and (b) after pile-up mitigation is applied. Each legend also reports the full-width half-max (FWHM) of a Gaussian fit to each of the projections, a smaller value being a stronger resolution.

Correlation showing the amount of pile-up energy density subtracted from a given trigger jet as a function of the matching, isolated offline reconstructed jet for -1.6 < η < 0.0. This was done on monte-carlo simulated tt̄ events with ⟨ μ ⟩ = 80 at a center-of-mass energy √s = SI14TeV. For offline jets below SI200GeV, there are usually not enough energetic glsplgTower around the seeded tower with E_T > SI15GeV to be included in the reconstruction and so those associated, lower energy trigger jets tend to be smaller in area as they have less towers and so the correction falls off with energy. At a certain point, the correction is approximately the same which is the ``full-occupancy" trigger jet with all glsplgTower within ΔR_cone < 1.0 of the seed participating in the reconstruction. For enough energy,

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange50100GeV to enhance hadronic W-bosons.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange50100GeV to enhance hadronic W-bosons.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange50100GeV to enhance hadronic W-bosons.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange100200GeV to enhance hadronic tops.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange100200GeV to enhance hadronic tops.

Trigger efficiency curves for the five different pile-up mitigation techniques. These are: no subtraction, no subtraction but a noise cut applied, no subtraction and simply shifted, with pile-up subtraction, and a hybrid cut. The legend reports the width of the turn-on w and the location of the plateau at 95% x_0.95. The efficiency is reported as a function of the large-R R=1.0 anti-k_t isolated offline jet matched to the given trigger jet for a trigger jet requirement of SI140GeV. An additional selection on the mass of the offline jet is required to be within SIrange100200GeV to enhance hadronic tops.

Correlations of the energy between leading gTower and the matched offline jet's leading subjet for (a) QCD multijet and (b) tt̄ monte-carlo simulated events with ⟨ μ ⟩ = 80 at a center-of-mass energy of √s = SI14TeV. The trigger jets are seeded with plgTower with E_T > SI15GeV and pile-up corrected using an upper threshold of SI6GeV. Notice that there is a nice average linearity and a relatively strong correlation (> 85%) for both monte-carlo samples. For multijet events, the leading gTower contains a majority of the energy of the offline jet, while for tt̄ this is lower as expected for jets with significant substructure and energy spread out more.

Correlations of the energy between leading gTower and the matched offline jet's leading subjet for (a) QCD multijet and (b) tt̄ monte-carlo simulated events with ⟨ μ ⟩ = 80 at a center-of-mass energy of √s = SI14TeV. The trigger jets are seeded with plgTower with E_T > SI15GeV and pile-up corrected using an upper threshold of SI6GeV. Notice that there is a nice average linearity and a relatively strong correlation (> 85%) for both monte-carlo samples. For multijet events, the leading gTower contains a majority of the energy of the offline jet, while for tt̄ this is lower as expected for jets with significant substructure and energy spread out more.

Distributions of the (a) gTower energy and (b) fraction of gFEX trigger jet energy carried by each of the leading towers in the given jet. The y-axis is exclusive binning in the number of subjets of the matched, isolated, reconstructed offline jet. Dashed lines connect points which map the same type of gTower to understand the trends of the leading, subleading, etc. plgTower as you require more and more substructure in the offline jet. Each distribution of gTower energies for an offline subjet selection is fitted to a Gaussian and the mean is extracted and drawn as the marker, while the standard deviation is extracted and drawn as error bars.

Distributions of the (a) gTower energy and (b) fraction of gFEX trigger jet energy carried by each of the leading towers in the given jet. The y-axis is exclusive binning in the number of subjets of the matched, isolated, reconstructed offline jet. Dashed lines connect points which map the same type of gTower to understand the trends of the leading, subleading, etc. plgTower as you require more and more substructure in the offline jet. Each distribution of gTower energies for an offline subjet selection is fitted to a Gaussian and the mean is extracted and drawn as the marker, while the standard deviation is extracted and drawn as error bars.

This is a plot of the (a) fake rate of dijet samples and (b) efficiency of the tt̄ samples for monte-carlo simulated events with ⟨ μ ⟩ = 80 and a center-of-mass energy √s = SI14TeV. (a) is the background fake rate where a gFEX trigger jet E_T selection is applied to the denominator and numerator and a ``subtower" multiplicity is applied to the numerator. (b) is the signal efficiency where the offline selection detailed in the plot is applied to the numerator and denominator and the trigger selection listed is applied to the numberator.

This is a plot of the (a) fake rate of dijet samples and (b) efficiency of the tt̄ samples for monte-carlo simulated events with ⟨ μ ⟩ = 80 and a center-of-mass energy √s = SI14TeV. (a) is the background fake rate where a gFEX trigger jet E_T selection is applied to the denominator and numerator and a ``subtower" multiplicity is applied to the numerator. (b) is the signal efficiency where the offline selection detailed in the plot is applied to the numerator and denominator and the trigger selection listed is applied to the numberator.

The signal efficiency is shown as a function of the background ``fake rate" from creffig:upgrade:substructure:eff:fakerate.

Overview of the ironman architecture.

Overview of the ironman server.

Overview of the ironman hardware interface.

Overview of the ironman client.

Overview of the ironman communications with the hardware.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in the boosted signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 0L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in boosted signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in moderate signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

N-1 plots for the analysis variables in compressed signal region for the Gtt 1L analysis.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Decay type the tt̄ component of the background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the dileptonic component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single-lepton component of the tt̄ background in the optimized Gtt-0L regions.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Composition of the single top component of the background in terms of Wt, t-channel and s-channel.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Heavy flavor composition of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the tt̄ component of the background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the semi-leptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.

[239] Breakdown of the dileptonic component of the tt̄ background in the optimized Gtt-1L regions.