Top-philic Machine Learning

TL;DR

This paper reviews modern machine learning techniques like CNNs, GNNs, and attention mechanisms, highlighting their applications in top quark searches at the LHC, including tagging, reconstruction, and inference methods.

Contribution

It provides a comprehensive overview of recent ML applications in top quark physics, emphasizing new approaches for event classification and likelihood-free inference.

Findings

Enhanced top tagging and reconstruction methods using ML

ML-based likelihood-free inference improves analysis accuracy

Generative models assist in unfolding and data simulation

Abstract

In this article, we review the application of modern machine-learning (ML) techniques to boost the search for processes involving the top quarks at the LHC. We revisit the formalism of Convolutional Neural Networks (CNNs), Graph Neural Networks (GNNs), and Attention Mechanisms. Based on recent studies, we explore their applications in designing improved top taggers, top reconstruction, and event classification tasks. We also examine the ML-based likelihood-free inference approach and generative unfolding models, focusing on their applications to scenarios involving top quarks.