# The Machine Learning Landscape of Top Taggers

## TL;DR

This paper compares various modern machine learning methods using low-level calorimeter data for top quark tagging, finding that despite architectural differences, their performance is similarly high, highlighting the power of these approaches.

## Contribution

It provides a comprehensive comparison of modern ML techniques for top quark tagging using low-level inputs, revealing their comparable effectiveness.

## Key findings

- Modern ML approaches perform similarly despite architectural differences.
- Low-level calorimeter data is highly effective for top tagging.
- New ML methods are powerful and promising for particle physics.

## Abstract

Based on the established task of identifying boosted, hadronically decaying top quarks, we compare a wide range of modern machine learning approaches. Unlike most established methods they rely on low-level input, for instance calorimeter output. While their network architectures are vastly different, their performance is comparatively similar. In general, we find that these new approaches are extremely powerful and great fun.