Playing Tag with ANN: Boosted Top Identification with Pattern Recognition

TL;DR

This paper introduces a novel top tagging algorithm using an Artificial Neural Network that treats calorimeter data as images, achieving high efficiency and low mis-tag rates in simulated LHC data.

Contribution

The paper presents a new pattern recognition-based top tagging method employing ANN on calorimeter images, improving performance over traditional algorithms.

Findings

Achieves 60% top-tag efficiency at 4% mis-tag rate for high p_T jets

Demonstrates the effectiveness of image-based pattern recognition in jet classification

Analyzes physical features and correlations with existing top-tagging observables

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.