Jet-Flavour Tagging at FCC-ee

TL;DR

This paper introduces advanced machine-learning algorithms for jet-flavour tagging at FCC-ee, significantly enhancing the identification of various quark and gluon jets to improve Higgs physics measurements.

Contribution

It presents novel machine-learning based jet-flavour identification algorithms that exploit particle-level data, improving discrimination of multiple jet types at FCC-ee.

Findings

High efficiency in b- and c-quark tagging

Ability to discriminate strange quark jets

Assessment of detector design impact on performance

Abstract

Jet-flavour identification algorithms are of paramount importance to maximise the physics potential of the Future Circular Collider (FCC). Out of the extensive FCC-ee physics program, flavour tagging is crucial for the Higgs physics program, given the dominance of hadronic decays of the Higgs boson. Highly efficient discrimination of $b$-, $c$-, $s$-, and gluon jets allows access to novel decay modes that cannot be identified at the LHC, adding quantitatively new dimensions to the Higgs physics programme. This contribution presents new jet flavour identification algorithms based on advanced machine-learning techniques that exploit particle-level information. Beyond an excellent performance of $b$- and $c$-quark tagging, they are also able to discriminate jets from strange quark hadronisation, opening the way to improve the sensitivity of the Higgs to strange quark coupling. The impact of different detector design assumptions on the flavour tagging performance is assessed using one of the baseline detector concepts for FCC-ee, IDEA.