Boosted top quark tagging and polarization measurement using machine learning

TL;DR

This paper explores the use of machine learning, specifically CNNs and BDTs, for improved boosted top quark tagging and polarization measurement, demonstrating enhanced sensitivity and better performance in leptonic decay channels.

Contribution

It introduces a CNN-based approach for top quark tagging and polarization measurement, outperforming traditional methods and providing a new measurable asymmetry variable.

Findings

CNN classifier is more sensitive to top quark polarization than standard variables

Leptonic channel yields better tagging performance and polarization sensitivity

Machine learning techniques improve top quark identification and polarization analysis

Abstract

Machine learning techniques are used for treating jets as images to explore the performance of boosted top quark tagging. Tagging performances are studied in both hadronic and leptonic channels of top quark decay, employing a convolutional neural network (CNN) based technique along with boosted decision trees (BDT). This computer vision approach is also applied to distinguish between left and right polarized top quarks. In this context, an experimentally measurable asymmetry variable is proposed to estimate the polarization. Results indicate that the CNN based classifier is more sensitive to top quark polarization than the standard kinematic variables. It is observed that the overall tagging performance in the leptonic channel is better than the hadronic case, and the former also serves as a better probe for studying polarization.