Please note that the name attached to each abstract is the one of the person submitting the abstract, and do not reflect the complete author list!
Recent Developments in Boosted Jet Tagging with Template Overlap Method
Mihailo Backovic (Weizman Institute)
Template Overlap Method (TOM) is a jet substructure tool aimed at matching the energy distribution of a fat jet to a partonic structure of a boosted decay. I will discuss the recent improvements to the previous formulations of the TOM algorithm, with an emphasis on the newly developed TOM formulation for boosted objects with missing energy. Our results show that the performance of TOM is weakly affected by pileup, even in the high luminosity environment characteristic of the LHC.
Jets without Jets
Daniele Bertolini (MIT)
Jets are key tools for physics at the LHC. Usually, jets are identified through a jet algorithm. In this talk, I will present an alternative way of thinking about jets, by showing how a broad class of inclusive jet-based observables can be replaced by event shapes. These event shapes do not require any jet clustering, but they still implement a jet-like pT cut on "jets" with an R-like radius. I will discuss various applications, including event selection at trigger-level, event-wide trimming, and alternative definitions for boosted objects identifiers.
Applying Computer Vision Techniques to Jet Flavor Identification:
From FisherFaces to FisherJets
Josh Cogan (SLAC)
We present a novel approach to identifying jet flavor, inspired by computer vision techniques. Calorimeter towers form the pixels of an image upon which we apply image analysis techniques to extract information from the shape of the jet's energy deposits. We construct the Fisher's Linear Discriminant and show its performance with respect to canonical jet identification techniques.
Divorcing Soft Substructure from Kinematics
David Curtin (Stony Brook)
Substructure analysis methods based on hard splittings, such as various taggers and jet groomers, have already found successful application at the LHC. The real-world utility of `soft' substructure, which is primarily sensitive to details of the parton shower, is much less well understood. We systematically investigate explicitly how much information is contained in these variables, divorced from hard kinematics. This is done in the context of a Wh->Wjj vs Wjj study, where the parton-level kinematics of all signal and background samples have been artificially made identical event-by-event. This ensures that correlations with kinematic observables do not fake a unique information content for soft substructure variables, and allows us to identify and devise variables which should receive the most experimental attention.
Learning how to count: a high multiplicity search for the LHC
Sonia El Hedri (SLAC)
We introduce a search technique that is sensitive to a broad class of signals with large final state multiplicities. Events are clustered into large radius jets and jet substructure techniques are used to count the number of sub-jets within each jet. The search consists of a cut on the total number of sub-jets in the event as well as the summed jet mass and missing energy. Two different techniques for counting sub-jets are described and expected sensitivities are presented for eight benchmark signals. These signals exhibit diverse phenomenology, including 2-step cascade decays direct three body decays, and multi-top final states. We find improved sensitivity to these signals as compared to previous high multiplicity searches as well as a reduced reliance on missing energy requirements. One benefit of this approach is that it allows for natural data driven estimates of the QCD background.
Jet Radiation Radius and Pileup
Zhenyu Han (Oregon)
I discuss the use of jet radiation patterns in the identification of boosted hadronically decaying massive particles. In order to minimize the contamination from pileups (as well as initial state radiation and the underlying event), it helps to evaluate jet substructure variables within small vicinities of the hard subjets. The optimal sizes of these areas are determined by the pt and the color structure of the jet. It is shown how to combine this observation with a subtraction scheme to improve W jet tagging in a highly contaminated environment.
Jet Substructure by Accident
Hou Keong Lou (Princeton)
We propose a new search strategy for high-multiplicity hadronic final states. When new particles are produced at threshold, the distribution of their decay products is approximately isotropic. If there are many partons in the final state, it is likely that several will be clustered into the same large-radius jet. The resulting jet exhibits substructure, even though the parent states are not boosted. This "accidental" substructure is a powerful discriminant against background because it is more pronounced for high-multiplicity signals than for QCD multijets. We demonstrate how to take advantage of accidental substructure to reduce backgrounds without relying on the presence of missing energy. As an example, we present the expected limits for several R-parity violating gluino decay topologies. This approach allows for the determination of QCD backgrounds using data-driven methods, which is crucial for the feasibility of any search that targets signatures with many jets and suppressed missing energy.
Jet Cleansing: Pileup Mitigation at High Luminosity
Matthew Low (U Chicago)
One of the greatest challenges in extracting useful information from high luminosity hadron-collision data is how to separate energy contributions from the primary interaction from those coming from secondary collisions. We propose a jet grooming technique called "cleansing" which can consistently correct for pileup on an event-by-event basis by exploiting the separation of track and calorimeter energy deposits in the data. By using track information to scale subjets, pileup can be actively removed in an observable-independent way. We present results for jet mass and the reconstruction of a dijet resonance.
QCD calculations for jet substructure
Simone Marziani (IPPP Durham)
We present first analytic calculations for commonly used substructure tools, namely jet mass distributions with trimming, pruning and the mass-drop taggers. These calculations lead to significant insight into the behaviour of the different taggers, especially in the very boosted regime. They also indicate how one can obtain variants of the above taggers that have better theoretical behaviour and that improve on signal/background discrimination.
Scale-invariant resonance tagging in multi-jet events
Juan Rojo (CERN)
We study resonant pair production of heavy particles in fully hadronic final states by means of jet substructure techniques. We propose a new resonance tagging strategy that smoothly interpolates between the highly boosted and fully resolved regimes, leading to uniform signal efficiencies and background rejection rates across a broad range of masses. Our method makes it possible to efficiently replace independent experimental searches, based on different final state topologies, with a single common analysis. As a case study, we apply our technique to pair production of Higgs bosons decaying into bbbar pairs in generic New Physics scenarios. We find that despite the overwhelming QCD background, the 4b final state has enough sensitivity to provide a complementary handle in searches for enhanced Higgs pair production at the LHC. We discuss the prospects for the application of similar ideas to other relevant cases, in particular merging the threshold and boosted topologies in BSM searches in fully hadronic ttbar production.
Exclusive Jet Mass Observables and Non-Global Logarithms
Robert Schabinger (U Mainz)
In this talk, we describe recent progress towards an analytical understanding of the soft non-global logarithms that appear in exclusive dijet mass observables. Non-global logarithms are a typical feature of sufficiently exclusive observables which arise from the presence of sharp boundaries between regions of phase space with different characteristic energy scales. Focusing on the so-called thrust cone jet algorithm in e+ e- annihilation, we exhibit explicit expressions for the non-global logarithms for arbitrary values of the jet cone size at leading two-loop order and then explain what simplifications become available in the limit of small R. We also show that, by making a judiciously chosen Lorentz boost, one can straightforwardly obtain the two-loop, small-R non-global logarithms by performing a simpler calculation using a hemisphere jet algorithm.
A top tagger based on shower deconstruction
Davison Soper (Oregon)
I present a method developed with Michael Spannowsky for tagging jets produced by hadronically decaying top quarks. The method is an application of shower deconstruction, an approach that we previously applied to identifying jets produced by Higgs boson decays. We tag an observed jet as a top jet based on a cut on a calculated variable chi. This variable is an approximation to the ratio of the likelihood that a top jet would have the structure of the observed jet to the likelihood that a non-top QCD jet would have this structure. I explain why this method is optimal among cut based methods to the extent that the likelihoods can be accurately calculated. I present some information about we calculate the approximate likelihoods. Finally, I present evidence that the shower deconstruction based tagger performs better than other publicly available tagging algorithms in discriminating boosted top quark jets from QCD jets using samples of events generated by Pythia.
Calculating Track-Based Observables for the LHC
Wouter Waalewijn (UCSD)
By using observables that only depend on charged particles (tracks), one can efficiently suppress pile-up contamination at the LHC. Such measurements are not infrared safe in perturbation theory, so any calculation of track-based observables must account for hadronization effects. I developed a formalism to perform these calculations in QCD, combining partonic cross sections with new non-perturbative objects called track functions. I will discuss properties of the track functions and illustrate with several examples how they can be used to calculate track-based observables.