Heavy Particle Jet Identification with Zest

TL;DR

The paper introduces 'zest', a new jet observable that effectively discriminates heavy particle jets from gluon jets, showing properties like boost invariance and stability, and compares favorably with machine learning methods.

Contribution

The paper proposes a novel jet observable called zest, analyzes its properties, generalizes it, and demonstrates its effectiveness in heavy particle jet discrimination, rivaling machine learning approaches.

Findings

Zest is boost invariant and stable against color exchanges.

Generalized zest can be optimized for improved discrimination.

Performance of generalized zest is comparable to machine learning top taggers.

Abstract

We introduce a new jet observable {\em zest} defined on exclusively constructed jets and study its potential to discriminate jets originated from Standard Model heavy particles like $W,~Z$ bosons and top quark from gluon initiated jets. Zest exhibits properties such as boost invariance, stability against global color exchange among partons, and inclusion or exclusion of a few soft particles in the jet. We also observe that for gluon jets, zest distribution is mostly insensitive to the jet mass. These properties make zest a suitable candidate for vetoing gluon jets at the colliders. Zest when used in conjunction with other substructure observables that are uncorrelated to it can further improve gluon jet veto. We generalize zest and show that in one limit it is synonymous to particle multiplicity and in the other limit, it projects only the leading particle. Optimization on the parameter of generalized zest further improves the discrimination ability of the observable. We find that for the top quark-initiated jets, the discrimination provided by generalized zest is in close comparison with a class of machine learning-based top taggers. We propose that studying other non-linear infrared and collinear unsafe observables may help in unveiling the hidden physics of machine learning-based observables.