Playing tag with ANN: boosted top identification with pattern recognition

Leandro G Almeida,Mathieu Cliche,Mihailo Backović,Seung J Lee,Maxim Perelstein

doi:10.1007/jhep07(2015)086

Abstract

Many searches for physics beyond the Standard Model at the Large Hadron Collider (LHC) rely on top tagging algorithms, which discriminate between boosted hadronic top quarks and the much more common jets initiated by light quarks and gluons. We note that the hadronic calorimeter (HCAL) effectively takes a “digital image” of each jet, with pixel intensities given by energy deposits in individual HCAL cells. Viewed in this way, top tagging becomes a canonical pattern recognition problem. With this motivation, we present a novel top tagging algorithm based on an Artificial Neural Network (ANN), one of the most popular approaches to pattern recognition. The ANN is trained on a large sample of boosted tops and light quark/gluon jets, and is then applied to independent test samples. The ANN tagger demonstrated excellent performance in a Monte Carlo study: for example, for jets with p T in the 1100-1200 GeV range, 60% top-tag efficiency can be achieved with a 4% mis-tag rate. We discuss the physical features of the jets identified by the ANN tagger as the most important for classification, as well as correlations between the ANN tagger and some of the familiar top-tagging observables and algorithms.

Highlights

Tensors [16, 17] and other perturbatively calculable jet shapes [18]
Many searches for physics beyond the Standard Model at the Large Hadron Collider (LHC) rely on top tagging algorithms, which discriminate between boosted hadronic top quarks and the much more common jets initiated by light quarks and gluons
We present a novel top tagging algorithm based on an Artificial Neural Network (ANN), one of the most popular approaches to pattern recognition

Summary

Event generation and pre-processing

We generate benchmark event samples with MadGraph 5 [44] at leading order, and shower them with Pythia 6 [45]. Rotate the coordinate system so that this principal axis is the same direction (+η) for all jets: ηj = ηj · cos(θ) + φj · sin(θ), φj = −ηj · sin(θ) + φj · cos(θ) These coordinate transformations remove information about the jet position in the calorimeter and its orientation in the (η, φ) plane. Both pieces of information are irrelevant for top tagging, and removing them from consideration allows the ANN tagger to focus on the irreducible physical differences between top and QCD jets.. In the language of image processing, each jet has been converted into an image with 30 ×30 pixels, with a grayscale color of each pixel given by the corresponding εab These images can be classified by an Artificial Neural Network (ANN), described

ANN tagger

Results

Discussion

A A brief description of top taggers used for benchmarking