Gaining (mutual) information about quark/gluon discrimination

By A. Larkoski, Jesse Thaler, W. Waalewijn. Journal of High Energy Physics, 2014

179 citations recorded by the source corpus.

Discriminating quark jets from gluon jets is an important but challenging problem in jet substructure. In this paper, we use the concept of mutual information to illuminate the physics of quark/gluon tagging. Ideal quark/gluon separation requires only one bit of truth information, so even if two discriminant variables are largely uncorrelated, they can still share the same “truth overlap”. Mutual information can be used to diagnose such situations, and thus determine which discriminant variables are redundant and which can be combined to improve performance. Using both parton showers and analytic resummation, we study a two-parameter family of generalized angularities, which includes familiar infrared and collinear (IRC) safe observables like thrust and broadening, as well as IRC unsafe variants like pTD and hadron multiplicity. At leading-logarithmic (LL) order, the bulk of these variables exhibit Casimir scaling, such that their truth overlap is a universal function of the color factor ratio CA/CF. Only at next-to-leading-logarithmic (NLL) order can one see a difference in quark/gluon performance. For the IRC safe angularities, we show that the quark/gluon performance can be improved by combining angularities with complementary angular exponents. Interestingly, LL order, NLL order, Pythia 8, and Herwig++ all exhibit similar correlations between observables, but there are significant differences in the predicted quark/gluon discrimination power. For the IRC unsafe angularities, we show that the mutual information can be calculated analytically with the help of a nonperturbative “weighted-energy function”, providing evidence for the complementarity of safe and unsafe observables for quark/gluon discrimination.

Subfields: Nuclear and High Energy Physics

View Paper: https://www.semanticscholar.org/paper/d3a7af3a3f50afabdf100f029df9377bca3cfa99

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Nuclear and High Energy Physics

Citations in the corpus

Energy flow polynomials: a complete linear basis for jet substructure (2017, background): “As already mentioned, EFPs are a special case of C-correlators [74], so not surprisingly, we find a close relationship between EFPs and other classes of observables that are themselves C-correlators, including jet mass, ECFs [51], certain generalized angularities [75], and energy distribution moment”
Les Houches 2017: Physics at TeV Colliders Standard Model Working Group Report (2017, background): “At the moment, joint distributions of two-point correlators are only known to NLL accuracy without grooming [657].”
Thinking outside the ROCs: Designing Decorrelated Taggers (DDT) for jet substructure (2016, background): “As has previously been noted, decorrelating variables which are not implicitly linearly correlated is poorly defined [54].”
Power counting to better jet observables (2014, background): “(2.6) 3The β = 1 minimized axes are also referred to as “broadening axes” [55, 59] as they correspond to axes that minimize the value of broadening [56, 64, 65].”
Classification without labels: learning from mixed samples in high energy physics (2017, background): “All of the observables can be written in terms of the generalized angularities [51] (see also [62–64]):”
Deep learning in color: towards automated quark/gluon jet discrimination (2016, background): “First-principles calculations of correlations among observables are also being explored [22].”
Looking Inside Jets (2019, background): “hapes have been introduced. In what follows, we review the most common ones. 5.4.1 Angularities and generalised angularities The simplest family of jet shapes is probably the generalised angularities [131] dened as = X i2jet z i Ri;jet R ; (5.11) where ziis the jet transverse momentum fraction ca”
Analytic boosted boson discrimination (2015, background): “In particular, for the (24)Differences between Monte Carlo generators for single emission observable can also be accentuated by departing from jet mass, and considering angularities, or energy correlation functions, or differences between quark and gluon jets, for which limited data from LEP can be ”
Les Houches 2015: Physics at TeV Colliders Standard Model Working Group Report (2016, methodology): “For general κ 6= 1, there are quasi-perturbative techniques based on generalized fragmentation functions [414] (see also [410,455–457]).”
Jet substructure at the Large Hadron Collider (2018, background): “Many of these observables exhibit Casimir scaling which results in nearly the same, limited discrimination power for all the observables [117, 131].”
Systematics of quark/gluon tagging (2017, methodology): “By contrast, because of the logarithms in the I(T ; Λ) integrand, one has to be careful about a potential binning bias [14].”
Energy flow networks: deep sets for particle jets (2018, background): “Observable O Map Φ Function F Mass m pμ F (xμ) = √ xxμ Multiplicity M 1 F (x) = x Track Mass mtrack p Itrack F (xμ) = √ xxμ Track Multiplicity Mtrack Itrack F (x) = x Jet Charge [72] Qκ (pT , Q pT ) F (x, y) = y/xκ Eventropy [74] z ln z (pT , pT ln pT ) F (x, y) = y/x− lnx Momentum Dispersion [93] p”
Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning (2017, background): “Particle multiplicity is among the most powerful observables for discrimination of quark and gluon jets [274, 275, 212], and further theoretical studies could provide understanding as to why it works so well and provide insight into the construction of new observables.”