# Jet grooming through reinforcement learning

## TL;DR

This paper presents a reinforcement learning-based jet grooming algorithm that optimizes jet properties for collider experiments, matching and potentially surpassing existing techniques in mass resolution for boosted objects.

## Contribution

The authors develop a deep reinforcement learning approach for jet grooming, introducing a modular, optimized method that adapts to various contexts and improves upon current state-of-the-art techniques.

## Key findings

- RL-based grooming matches state-of-the-art performance
- Improves mass resolution for boosted objects
- Provides a flexible, modular grooming framework

## Abstract

We introduce a novel implementation of a reinforcement learning (RL) algorithm which is designed to find an optimal jet grooming strategy, a critical tool for collider experiments. The RL agent is trained with a reward function constructed to optimize the resulting jet properties, using both signal and background samples in a simultaneous multi-level training. We show that the grooming algorithm derived from the deep RL agent can match state-of-the-art techniques used at the Large Hadron Collider, resulting in improved mass resolution for boosted objects. Given a suitable reward function, the agent learns how to train a policy which optimally removes soft wide-angle radiation, allowing for a modular grooming technique that can be applied in a wide range of contexts. These results are accessible through the corresponding GroomRL framework.