Presentation materials
I will discuss fractal jet observables, which are collinear-unsafe but can be described by generalizing the formalism of fragmentation functions. Generalized fragmentation functions (GFFs) are nonperturbative objects with a calculable RG running. In contrast to the linear DGLAP equations for ordinary fragmentation functions, GFFs evolve nonlinearly, since they encode correlations among subsets...
In this talk, we introduce a new jet substructure method based on a
recursive iteration of the Soft Drop algorithm through both branches
of the clustering tree.
Recursive soft drop uses an additional parameter N to define the
number of layers of soft drop declustering, providing an optimized
grooming strategy for boosted objects with (N+1)-prong decays, as well
as improved stability in high...
Charged track multiplicity is among the most powerful observables for discriminating quark- from gluon-initiated jets. Despite its utility, it is not infrared and collinear (IRC) safe, so perturbative calculations are limited to studying the energy evolution of multiplicity moments. While IRC-safe observables, like jet mass, are perturbatively calculable, their distributions often exhibit...
Distinguishing quark-initiated from gluon-initiated jets is useful for many measurements and searches at the LHC. We present a quark-initiated versus gluon-initiated jet tagger from the ATLAS experiment using the number of reconstructed charged particles inside the jet. The measurement of the charged-particle multiplicity inside jets from Run 1 is used to derive uncertainties on the tagger...
Distinguishing between quark and gluon initiated jets relies on differences in the QCD shower patterns and is an important ingredient for a number of physics analyses. We present the current status of quark gluon tagging in CMS including comparisons using 13 TeV collision data.
Small radius jets with R = 0.4 are standard tools in ATLAS for physics analysis. They are calibrated using a sequence of Monte Carlo simulation-derived calibrations and corrections followed by in-situ calibrations based on the transverse momentum balance between the probed jets and well-measured reference signals. In this talk the inputs to jet reconstruction in LHC Run 2 comprising...
Measurement of jet energy scale corrections and resolution, and performance of jet mass scale and resolution based on data collected at a center-of-mass energy of 13 TeV are presented in this report. Jet energy scale corrections at CMS accounts for the effects of pileup, and dependencies of response of jets on transverse momenta and detector non-uniformity. The differences in response measured...
Large-R jets are used by many ATLAS analyses working in boosted regimes. ATLAS Large-R jets are reconstructed from locally calibrated calorimeter topoclusters with the Anti-k_{t} algorithm with radius parameter R=1.0, and then groomed to remove pile-up with the trimming algorithm with f_{cut} 0.05 and subjet radius R=0.2. Monte Carlo based energy and mass calibrations correct the...
The case of CMS "two prong" tagging algorithms are presented, specifically detailing the cases of W and H boson tagging. This talk will focus on the most recent algorithms used in LHC Run II analyses and their validation in data.
We present updates of W, Top and Higgs tagging studies with the ATLAS detector. The performance of 2 variable taggers, HEPTopTagger and shower deconstruction are compared in Monte Carlo simulations. To asses the modelling of the taggers’ performance, the tagging efficiencies are measured, with the full 2015+2016 dataset, in semi-leptonic top quark pair events and the background rejections are...
An overview of methods for identifying decays of boosted top quarks with the CMS detector in Run II is presented.
The performance of standard tagging algorithms begins to fall in the case of highly boosted B hadrons (γβ=p/m>200). This work builds on our previous study that uses the jump in hit multiplicity among the pixel layers of an ATLAS or CMS-like detector when a B hadron decays within the detector volume. Consequently, tracking is not required.
First, multiple pp interactions within a finite...
Machine learning based on convolutional neural networks can be used to study jet images from the LHC. Top tagging in fat jets offers a well-defined framework to establish our DeepTop approach and compare its performance to QCD-based top taggers. We optimize a network architecture to identify top quarks in Monte Carlo simulations of the Standard Model production channel. Using standard fat...
We present techniques for the identification of hadronically-decaying W bosons and top quarks using high-level features as inputs to boosted decision trees and deep neural networks in the ATLAS experiment at sqrt(s)=13 TeV. The performance of these machine learning based taggers is compared in Monte Carlo simulation with various different tagging algorithms. An improvement in background...
Machine learning has become an important tool in particle physics, and in jet substructure and boosted objects in particular. This presentation shows the breadth of applications from CMS, from "DeepFlavor" b-tagging to new techniques of substructure applications.
We develop a new pileup mitigation technique based on multi-channel jet images using convolutional neural nets. The input to the network is a three-channel jet image: the calorimeter "pixel" information of charged leading vertex particles, charged pileup particles, and neutral particles . We compare our algorithm to existing methods on a wide range of simple and complex jet observables up to...
Pileup is one of the biggest challenges facing the LHC and HL-LHC physics programs. Many reconstruction methods have been proposed for mitigating its effects across a broad range of physics metrics such as jet and jet substructure response and resolution, missing transverse energy performance, and lepton identification. Among the most successful are the SoftKiller and Pileup Per Particle...
Simultaneous proton-proton collisions, or pileup, at the LHC has a significant impact on jet reconstruction, requiring the use of advanced pileup mitigation techniques. Pileup mitigation may occur at several stages of the reconstruction process, and ATLAS uses a combination of schemes, including constituent reconstruction methods, constituent-level pileup-mitigation techniques, and jet-level...
We present tools developed by CMS for LHC Run II designed for pileup mitigation in the context of jets, MET, lepton isolation, and substructure tagging variables. Pileup mitigation techniques of "Pileup per particle ID" (PUPPI), and pileup jet identification are presented in detail along with the validation in data.
Since the machine learning techniques are improving rapidly, it has been shown that the image recognition technique can be used to detect jet substructure. And it turns out that deep neural networks can match or outperform traditional approach. To push it further, we investigate the Recursive Neural Networks (RecNN), which embeds jet clustering history recursively as in natural language...
