Abstract
A large number of physics processes as seen by the ATLAS experiment manifest as collimated, hadronic sprays of particles known as ‘jets.’ Jets originating from the hadronic decay of massive particles are commonly used in searches for new physics. ATLAS has employed multivariate discriminants for the challenging task of identifying the origin of a given jet. However, such classifiers exhibit strong non-linear correlations with the invariant mass of the jet, complicating analyses which make use of the mass spectrum. A comprehensive study of different mass-decorrelation techniques is performed with ATLAS simulated datasets, comparing designed decorrelated taggers (DDT), fixed-efficiency k-NN regression, convolved substructure (CSS), adversarial neural networks (ANNs), and adaptive boosting for uniform efficiency (uBoost). Performance is evaluated using suitable metrics for classification and mass-decorrelation.
Highlights
Jets are used to reconstruct hadronic decays of massive particles, e.g. W and Z bosons, in several searches in the ATLAS experiment [1] at the Large Hadron Collider (LHC)
multivariate analysis (MVA) taggers using only jet substructure observables as inputs yield tagger observables which exhibit non-linear correlations with the jet mass
This means that simple selections on such MVA tagger observables tend to distort the non-resonant background jet mass distribution, sculpting it to resemble the resonance jet mass peak, and thereby complicating resonance searches in the jet mass spectrum
Summary
Jets are used to reconstruct hadronic decays of massive particles, e.g. W and Z bosons, in several searches in the ATLAS experiment [1] at the Large Hadron Collider (LHC). Fixed-efficiency k-nearest neighbours (k-NN) regression can be considered a non-parametric generalisation of DDT In this contribution, the method is used to construct a new substructure observable based on D2, which, for the background process, is decorrelated from the jet mass and pT. The two networks are trained simultaneously with gradient reversal, according to Eq (5), with a small learning rate for the classifier relative to the adversary, resulting in the zANN observable During training, both signal and background samples are re-weighted to have flat pT distributions. The second MVA-based mass-decorrelation method builds on the AdaBoost method for boosting decision updated tarseews,it+wth=erwe itat×ecvtie,rwy hbeoroestthinegcslatespsifitctahteiornelwateiivgehtwcetiigihstbwasiteodf each training example i is on whether the ith sample was misclassified by the decision tree at step t This results in a jet tagger zAdaBoost ∈ [0, 1]. Both BDT taggers use the same substructure variables and sample re-weighting during training as the (A)NN taggers
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
