CERN Accelerating science

Talk
Title	GroomRL: jet grooming through reinforcement learning
Video	If you experience any problem watching the video, click the download button below Download Embed Show n. of views Mp4: Medium (1000 kbps) High (4000 kbps) More.. 800 kbps 800 kbps 800 kbps 1000 kbps 1000 kbps 2000 kbps 2000 kbps 2000 kbps 4000 kbps 4000 kbps Copy-paste this code into your page: <iframe width="640" height="360" frameborder="0" src="https://cds.cern.ch/video/2672625?showTitle=true" allowfullscreen></iframe> Copy-paste this code into your page to include both slides and lecture: <iframe src="https://mediastream.cern.ch/MediaArchive/Video/Public2/weblecture-player/index.html?year=2019&lecture=766872c25" width="1020px" height="600px" allowfullscreen scrolling="no" frameborder="0"></iframe>
Author(s)	Dreyer, Frederic Alexandre (speaker) (Oxford)
Corporate author(s)	CERN. Geneva
Imprint	2019-04-17. - 0:20:33.
Series	(LPCC Workshops) (3rd IML Machine Learning Workshop)
Lecture note	on 2019-04-17T16:30:00
Subject category	LPCC Workshops
Abstract	We introduce a novel implementation of a reinforcement learning algorithm which is adapted to the problem of jet grooming, a crucial component of jet physics at hadron colliders. We show that the grooming policies trained using a Deep Q-Network model outperform state-of-the-art tools used at the LHC such as Recursive Soft Drop, allowing for improved resolution of the mass of boosted objects. The algorithm learns how to optimally remove soft wide-angle radiation, allowing for a modular jet grooming tool that can be applied in a wide range of contexts.
Copyright/License	© 2019-2024 CERN
Submitted by	paul.seyfert@cern.ch