libGroomRL: Reinforcement Learning for Jets

TL;DR

This paper introduces libGroomRL, a C++ library that employs deep reinforcement learning to develop jet grooming algorithms, achieving comparable or better performance than existing methods at the LHC by optimizing jet mass resolution.

Contribution

The paper presents a novel reinforcement learning-based approach for jet grooming, providing a modular C++ library that integrates with FastJet for improved jet analysis.

Findings

RL-trained grooming matches state-of-the-art techniques

Improves mass resolution for boosted objects

Flexible, modular grooming method

Abstract

In these proceedings, we present a library allowing for straightforward calls in C++ to jet grooming algorithms trained with deep reinforcement learning. The RL agent is trained with a reward function constructed to optimize the groomed 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. The neural network trained with GroomRL can be used in a FastJet analysis through the libGroomRL C++ library.