BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Prediction of electromagnetic fraction in a hadronic shower using 
 deep neural network
DTSTART;VALUE=DATE-TIME:20221201T143000Z
DTEND;VALUE=DATE-TIME:20221201T144500Z
DTSTAMP;VALUE=DATE-TIME:20260719T180201Z
UID:indico-contribution-3136@cern.ch
DESCRIPTION:Speakers: Marina Chadeeva (P.N. Lebedev Physical Institute of 
 RAS (LPI))\nThe intrinsically large variation of the energy deposited in a
  calorimeter by hadrons imposes limitations on the improvement of hadron e
 nergy resolution. The fluctuation of electromagnetic fraction within a had
 ronic shower is known to be one of the main sources of such variations. Se
 veral techniques were developed to improve the energy resolution for hadro
 ns including the so called hardware compensation (compensating and dual-re
 adout calorimeters) and software compensation approaches. The reliable pre
 diction of the amount of electromagnetic fraction on an event-by-event bas
 is opens a possibility to correct the energy during the offline reconstruc
 tion and improve the energy resolution. In this study\, the samples were i
 nvestigated of hadronic showers simulated with physics lists from Geant4 p
 ackage version 10.3 in the model of a highly granular hadron calorimeter f
 or the initial hadron energies 10--80 GeV. The deep neural network was tra
 ined using a supervised learning and calorimetric observables as inputs to
  predict the electromagnetic fraction in a shower. The achieved neural net
 work performance and observed improvement in hadron energy resolution of m
 ore than 15% are presented and discussed.\n\nhttps://indico.particle.mephi
 .ru/event/275/contributions/3136/
LOCATION:Hotel Intourist Kolomenskoye 4* Moskvorechye 1
URL:https://indico.particle.mephi.ru/event/275/contributions/3136/
END:VEVENT
END:VCALENDAR