BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning from the Lund plane - Frederic Dreyer (University of Oxfo
 rd)
DTSTART:20190528T150000Z
DTEND:20190528T160000Z
UID:TALK125596@talks.cam.ac.uk
CONTACT:67735
DESCRIPTION:Lund diagrams\, a theoretical representation of the phase spac
 e within jets\, have long been used in discussing parton showers and resum
 mations. I will show that they can be created for individual jets through 
 repeated Cambridge/Aachen declustering\, providing a powerful visual repre
 sentation of the radiation within any given jet. Concentrating on the prim
 ary Lund plane\, I will outline some of its analytical properties\, highli
 ght its scope for constraining Monte Carlo simulations and comment on its 
 relation with existing observables. I will then examine its use for booste
 d electroweak boson tagging\, showing that it can provide good performance
  when used as input to machine learning approaches or within a log-likelih
 ood method. Finally\, I will discuss applications to the issue of jet groo
 ming\, and introduce a framework to automate the definition of a jet groom
 ing algorithm through the use of reinforcement learning\, showing how the 
 removal of soft wide-angle partons can be optimized by the RL agent throug
 h an appropriate choice of reward function.
LOCATION:Ryle Seminar Room (Rutherford 930) 
END:VEVENT
END:VCALENDAR